Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: What are you working on? (July 2025)

news.ycombinator.com

Post reply on HN
Print view View original

871–880 of 900 posts

Re: Ask HN: What are you working on? (July 2025)

#871

Post by vb-8448 »

danfunk wrote: post #853
vb-8448 wrote: post #556

A workflow engine, but not an "airflow like" (even if I hope it will become an airflow killer) but more similar to old job scheduling systems from mainframe era(OPC/TWS/CA7 ecc). In my opinion, those mainframe-era dinosaurs are conceptually far more superior to what we have today on servers and cloud systems.

We are working on a workflow engine as well - picking up concepts from the past - not quite as far back as mainframes, but I respect looking back to the past for things that work well, and re-applying them to a new world. I'd be up for sharing notes. Our work is here https://github.com/sartography/spiff-arena -- live demo is here https://spiff.works/agent-demo

Hi, thank you for your feedback.

I think in the past, I've already had a look at your repo, but it's a little bit different from my idea. Your is build around BPMN (and, if I'm not wrong, it's basically a easy way to create custom finite state machines) while I'm working on a "lower level".

I have 3 concepts: tasks, links and workflows(a collection of tasks). Links can be created between task of the same workflow or task from different workflows. Workflows can be scheduled based on calendar rules(cron like or "2nd working day of the month" and similar) or external events(api call, new file on s3, ecc ecc). Normally workflows are DAGs, but loops can be created if needed.

The other idea is that it's just the orchestration layer, the task needs to be executed by specific "agent", that can be written in any language (there is a very simple protocol build on top of http to get the tasks from the orchestrator or to send the results) and can be installed on local or remote server.

Just some context: in the mainframe era, businesses used to be orchestrated by a unique "scheduler" running all day long(it was so flexible that basically any business from banking to insurance to manufacturing to industry were able to adapt it to their needs). Bigger companies used to have >100k (or even millions) of tasks per day with very well-defined dependencies. The "good" part was that it was very easy to monitor and manage your batch operations. My idea is to reproduce something similar, but also to add a better support for distributed envs and "human" tasks.

Re: Ask HN: What are you working on? (July 2025)

#872

Post by wczerniak »

I'm finishing Flatcal https://flatcal.com a service that merges icalendar streams from various sources into one. It's weird that we can create multiple calendars to better organize scheduling and then to share availability we have to send each one as a separate link. It helps with more than that but sharing multiple calendars as one is the main goal atm.

Re: Ask HN: What are you working on? (July 2025)

#874

Post by anhldbk »

01HNNWZ0MV43FF wrote: post #822
anhldbk wrote: post #450

I'm working on Beam, a web app that lets you transfer files between nearby devices using QR codes - completely offline, with no servers involved, and leaving absolutely no trace. Beam is perfect for sharing sensitive documents, transferring files when you can't use USB, email, or cloud storage. Try it here: https://get-beam.vercel.app

Bluetooth, or animated QR?

No bluetooth. Just use cameras & screens.

Re: Ask HN: What are you working on? (July 2025)

#875

Post by Breza »

alextousss wrote: post #724
teruza wrote: post #681

Earlier quoted context omitted.

Very cool idea. What is your estimated price? This could work well in many African countries if the price is low.

Long term, around $300/unit

If it worked in my DC backyard, I'd pay that in a heartbeat.

Re: Ask HN: What are you working on? (July 2025)

#876

Post by warpech »

davidpolberger wrote: post #856
junto wrote: post #839

Earlier quoted context omitted.

Is this like gorules?

Other apt comparisons to what I'm building include HyperFormula and Microsoft Power Fx.

As a HyperFormula founder, thank you for the mention. I wonder what’s your opinion on our formula library

Re: Ask HN: What are you working on? (July 2025)

#877

Post by nymanjon »

I built a simple lib for JS/HTML called html-traits[1]. It's similar to built-in web components but instead of using the `is` attribute you use the `traits` attribute. So, it is specifically meant for progressive enhancement. Let's say I have a `textarea` element and I want it to have the behavior where the text box expands automatically, I would add a trait called "expando" that would do that for me. But let's say I also want it to turn red when there's more than 200 characters entered, so I could add another trait called "character-limit" with an added attribute `data-character-limit="200"`.

So, the element works like normal if no JS is enabled or if the JS broke for some reason, but it has those extra behaviors when JS is working. And it is very clean as it is just attributes on the element rather than having nested element hell like when you are using normal web components, or worse, like when you have web components which don't even show the `textarea` element at all and they put it in the shadow DOM and the developer has to reimplement all the behavior of the original element.

So, composition over inheritance.

[1]: https://github.com/jon49/html-traits

Re: Ask HN: What are you working on? (July 2025)

#878

Post by bpavuk »

antfarm wrote: post #562
bpavuk wrote: post #485

ATProto is helluva complex and inconvenient to work with without supporting libraries, even in a relatively sugary language, like Kotlin. So, I decided to write a (programmatic) code generator that takes Lexicon on input and gives all the HTTP code and data classes you may ever need. If you wonder why Bluesky native apps are crap, that's because they are in TypeScript, not in Kotlin and Swift, which are programming l…

Can you elaborate on BlueSky native apps being bad because they are written in TypeScript?

TypeScript does not compile to machine code (in case of Android, ART bytecode) directly. Instead, it compiles to JS, which has a lot of well-known problems and is not as performant as machine code. Yeah, iPhones not older than 3-5 years will not notice it because of Apple's best-in-class flagship chipsets, but hey, 1) not everyone in the world has an iPhone and 2) why waste computing power and battery life? WASM wouldn't exist if not for JS's poor performance - it simply would not need to! I'm still glad TS exists, though, because it fixes quite a lot of problems with JavaScript DX just by marrying it with types. The only problem of TypeScript is JavaScript. (Arguably, also Microsoft, but this is a separate discussion.) Heck, the biggest problem of the whole web IS JavaScript - that's why there is a comeback trend in HTML/CSS-centric frameworks. JS is simply unable to sustain such heavyweight frameworks as React (including React Native, which is what Bluesky uses). Such frameworks as SwiftUI and Compose are living proofs of that - while you stay with Kotlin/Native or Kotlin/JVM targets (iOS, Android, JVM desktop), your app feels snappy, assuming you wrote it well, because you stay close to the metal. Once you move to Kotlin/JS, though, you are fucked. Immediately, you lose tons of performance, making such a beautiful framework with the best animations API in the world usable only for PowerPoint-style slideshows. Kotlin/WASM backend is still experimental, but future Kotlin/JS backend deprecation is confirmed.

Re: Ask HN: What are you working on? (July 2025)

#880

Post by swamp-agr »

Anti-spam bot plugin for messengers:

- MVP version for Telegram (since spamming is a part of their business model, it feels natural to start with it)

- More precisely, data pipeline for weights and measurements for word frequencies. Think of it as small-language models.

- More precisely, it is about morphological analysis of words across different languages. Unlike Meta with regex-based dictionaries[0], I am porting rule-based morphology analysis python library[1] into target programming language.

- More precisely, right now it is about understanding DAWG data structure by porting it from C++[2] to Haskell[3].

- Instead of introducing FFI I wanted to become more comfortable with LLMs, I am trying to approach their internals (or my possibly wrong vision of their internals) by building small language model based on a corpus of thousands of spam messages.

Links:

[0] duckling: https://hackage.haskell.org/package/duckling

[1] pymorphy2: https://github.com/pymorphy2/pymorphy2

[2] dawgdic/C++: https://code.google.com/archive/p/dawgdic/

[3] dawgdic/Haskell (work-in-progress): https://github.com/swamp-agr/dawgdic

Post reply on HN