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

I got filtered by the Ent arc of LOTR and dropped the book.

Not if tree cutting simulator is a VR experience

You can start right now with an algorithm I learned from an expert when I was working in a landscaping business.

It is a very simple three-pass plan: "Deadwood, Crossovers, Aesthetics".

So, first pass, go through the tree cutting out only and all the dead branches. Cut back to live stock, and as always make good clean angle cuts at a proper angle (many horticulture books will provide far better instructions on this).

Second pass, look only for branches that cross-over other branches and especially those that show rubbing or friction marks against other branches. Cut the ones that are either least healthy or grow in the craziest direction (i.e., crazy away from the normal more-or-less not radially away from the trunk).

Then, and only after the other two passes are complete, start pruning for the desired look and/or size & shape for planned growth or bearing fruit.

This method is simple and saves a LOT of ruined trees from trying to first cut to size and appearance, then by the time the deadwood and crossovers are taken later, it is a scraggly mess that takes years to grow back. And it even works well for novices, as long as they pay attention.

I'd suspect entering the state and direction of every branch to an app would take longer than just pruning with the above method, although for trees that haven't fully leafed out, perhaps a 360° angle set of drone pics could make an adequate 3D model to use for planning?

In any case, good luck with your fruit trees — may they grow healthy and provide you with great bounty for many years!

When i read OP this is what I thought it was going to be - these branches are going to be apex competitors, these are crossing or going to cross, this one shows signs of disease, this one interrupts air flow through the centre, etc.

Or perhaps follows the steps correctly.

I was thinking I could use a tool just like the wire you described to remove a stump, after I spent 6 hours with a 5ton Bobcat trying to dig up a 3ft diameter pine stump to no avail today. For felling trees though, you need precise front cuts/back cuts to drop the tree at a desired angle, you can't just cut in one direction even if you have a cable attached.

Perhaps drilling in a wedge shape so it weakens the branch and it eventually breaks off naturally but it seems like more work than just a chain saw. The holes could also be used for steam treatment, enzymes [0] or something else to break it.

[0] https://en.wikipedia.org/wiki/Lignin-modifying_enzyme

I use a tool called the Alligator. Is is a tool you can use like scissors. It has 2 chains on the inside of the the business ends. You put around the branch. Close the end and press the button. Springs will then close the end even more and cut the wood. No open chains

I hear you on the safety angle, but chainsaws work really well. They're also very versatile.

We use a winch to guide the branches down, but would never apply the winch directly to the tree in case of whiplash when the branch finally breaks

Yeah, I agree. It is a pain to search product by product instead of sticking to one store. Also popgot.com can only do what's online & shipped to you -- so really just the non-perishables / daily essentials that are not fresh groceries. But even when limited to consumables I save ~$100/mo by basically buying by unit price.

Uploading a receipt to see how much you can save... that's a good idea. I think I can find your email via your personal site. Can I email you when we have a prototype ready?

This is a great idea. And OCR should be good enough nowadays to parse the receipts. Probably would work best as a mobile app, though.

My pleasure! Happy you could use it as much as I do. Anyway we can chat in person? I'd love to make more stuff for you. chris@<our site>.com

> For instance, people often tell me Costco is much cheaper than alternatives, and for me to compare I have to compile my shopping cart in multiple stores to compare.

A few years ago, I was very diligently tracking _all_ my family's grocery purchases. I kept every receipt, entered it into a spreadsheet, added categories (eg, dairy, meat), and calculated a normalized cost per unit (eg, $/gallon for milk, $/dozen eggs).

I learned a lot from that, and I think I saved our family a decent amount of money, but man it was a lot of work.

I'm so glad you like it!

We have historical price tracking in the database, but haven't exposed it as a product yet. What do you have in mind / what would you use it for?

Ah, hell yeah! My buddy on this project has been itching to add sweetmarias.com ... he just needed this as an excuse.

So yeah, we'll add it. If you shoot me an email (or post it here?) to chris @ <our site>.com I'll send you a link when it's done. Should take a day or two.

Thank you, nice to get some idea validation from folks in the industry. For sure data warehouses are the top priority on my TODO list, I picked PG first because that's what I'm familiar with.

I can ping you via email when the debugger is ready, if you're interested. My email is in my profile

Can you not use EXPLAIN ANALYZE to identify steps that had the highest compute time? I think most databases have some form of this.

You're onto the original idea I started out with! Unfortunately it's very difficult to correlate input SQL to an output query plan – but possible. It's definitely in future plans

MSSQL has the execution plan thing that will tell you which steps are involved and how long they take.

Thanks for the suggestion! I am using https://github.com/tobymao/sqlglot, which magically supports most SQL dialects. And yes, support for DuckDB is also in future plans

Thanks! Would you mind sharing what would be your use cases?

At my job, all of our business logic (4 KLOC of network topology algorithms) is written in a niche query language, which we have been migrating to PostgreSQL. When an inconsistency/error is found, tracking it can take days, manually commenting out parts of query and looking at the results.

Thanks for the motivation to finish this as soon as possible :) I'm working on a basic landing page with screenshots/videos and a "get notified" button right now – shoot me an (empty, if you want) email (in profile) and I'll ping you as soon as it's ready

https://dequery.io :) Added a little signup form (with possibility of providing additional feedback)

What database are you using? I'd be happy to hear about your usecases and hopefully help you, shoot me an email (in profile)

I've never really dabbled in youtube. I have several projects/papers I am working on using this code, I have thought about writing some blog posts as I publish those. But a PhD is going to be a major time sink, we will see what happens.

That's the first thing I'm going to do as soon as it's possible! I recently refactored the code base to a monorepo, and still need to make some adjustments so it'll run stable again. Stay tuned :)

Ebooks in an S3 compatible storage bucket, metadata in a Postgres database. That has the huge advantage of being able to do full text search and kNN similarity right in the database, for example.

Colibri is built around a pretty solid data schema (I hope). Check out the migrations folder if you’re curious :-)

Thanks you, that means a lot. And good luck for your search!

So, Colibri is intended to be a companion to Calibre, maybe do like 80% of what it does, but not all of it. Also, I want Colibri's core to have a fully clean collar: It handles your personal books, but will not be able to automatically de-DRM books and such. However, the way book assets work, you're of course free to just attach an encrypted AZW3 file to a book!

> Are these ebooks supported and does it only supports metadata, what about the content search for these ebooks?

Have you seen https://github.com/colibri-hq/colibri/issues/45? Content search is planned, but requires access to a book's text content, obviously. My recommendation would be to use Calibre to strip DRM and convert the books to epub/mobi files, and import those to Colibri; this has the general benefit of ensuring access to content you bought without depending on Amazon's good will :-)

All types. The sugar dashboard allows import of data from different glucose apps, so its goal is to allow you visualize and analyze your data. I hope to integrate with cgms directly if I get some that allow it, and also source from Health connect. Sharing with specific people eg doctor is also a big ask that I'm working on. The other WIP tools will be fore general health, not just diabetes, like carb counting from a photo via AI

yes, came across xdrip+ when looking for an android app I could use for Libre 2. I don't think Dexcoms are sold in Kenya, and even the Libres around are UK ones so you need 1) a VPN to setup, 2) an iphone. Both things being a challenge for most - I had to buy a my first ever iphone for this. Anyway, found xdrip a bit of a challenge to setup and a bit too technical to suggest to others; needs sideload and manually disabling a lot of Android defaults.

I had a lot of success with Juggluco[1] which is available on the Play Store and provides easy to use APIs to interact with supported CGM readings. Juggluco has an inbuilt xdrip web server but I haven't tried it yet.

Will definitely look into xdrip+ further.

[1] https://github.com/j-kaltes/Juggluco

Thanks! I started out with a Nextjs full stack on Vercel, with db on Turso but ended up with a React frontend (next on vercel) and Go backend (selfhosted on vps).

Decided to port the backend to Go + postgres (on a Hetzner VPS), and retain the frontend on Nextjs - A lighter weight client, moving most of the compute to the backend API. Few reasons for the port: I've had a lot more success/stability with Go backends, Turso pulled multi-tenant dbs which is what I mostly wanted them for, Nextjs is getting too hard for me.

Go backend is just the std lib (1.22+ server with the nice routing) - I mostly write all the lines in this

Frontend is textbook modern react: React19,next15,tailwind4 - AI mostly writes the code in the frontend (Cursor + Cline + sequentialthinking + context7 + my own custom "memory bank" process of breaking down tasks). AI is really, really good at this. I wrote this https://image-assets.etelej.com/ in literally 2 days 2 weekends ago with less than 10% of code being mine (mostly infra + hono APIs)

Thanks. It is not sexy. There also many YouTube building channels. They concentrate on nice content while I live in construction site and would love to finish it asap. It’s also not really interesting topic - nobody builds brick houses anymore. Mortar was replaced with glue and bricks with much bigger building blocks.

> Undershooting means paying for someone's product to make up for the difference

Indeed. Making up the difference can easily eat most of the farmer's profits. I guess it is even more pronounced for berries when compared to grains, because they cannot be stored for so long.

Interesting. Thanks!

It is possible. However, getting accurate yield data requires a "smart" harvester that can produce yield maps. Many modern harvesters are equipped with GPS and various sensors, so it is possible. However, farmers are really slow to replace old equipment if it works fine. I guess there are some retrofit solutions for yield mapping, but I haven't investigated their affordability and penetration into the (EU) farming landscape yet. Additionally, there are other interesting parameters apart from the harvested quantity that can be captured (e.g. the quality of the grain itself, such as size, composition, humidity etc).

Fisher invented ANOVA specifically for analyzing crop yields so it's a natural fit.

However for precision agriculture kavalg might want to consider other methods.

I am not sure that I understand your question correctly, but given more precise sunflower density estimation, the farmer has three options:

1. Plow the field and seed again (same or different variety or grain). This is a very crude measure, but it is sometimes the right thing to do, because as I said most of the expenses have not been realized yet (fertilizer, pesticide, fuel, payroll, paying rent for the land). It is also a time critical decision, because the window of opportunity for plowing and reseeding is not very wide.

2. Accept the lower yield if it is within a reasonable margin (e.g. comparable to the expenses to plow and reseed).

3. Do partial reseeding over the existing plants (without plowing). This is an emerging strategy with the proliferation of smart seeders, but it requires a precise seeding map to be created beforehand (i.e. based on the density estimate). As an advantage, you spare the expenses for seeds and plowing, however there is some disadvantage as well, due to the different rate of development of the newly seeded plants. Farmers usually need plants to be ready for harvest at the same time, otherwise the quality of the grains suffers and hence the selling price is lower.

In addition to these points, having precise density information after germination helps with the identification of problems, such as seeder malfunction (e.g. nozzles getting clogged), seed quality and meteo data (e.g. too much rain, low temperatures etc).

Hmm that's strange. Perhaps you have two Google accounts with different subscriptions? When you sign in I grab your subscribed channels and that's what I send when you categorise. You can check the "Subscriptions" page to see exactly what data we pulled.

I did say Chrome browser because with the deprecation of manifest v2, I had to make a choice about which to support. I decided given Chrome's larger market share that it would benefit the most people sooner. However I'm building it in such a way that porting to Firefox shouldn't require much additional work.

Yeah absolutely! The category list is meant to be dynamic as the industry changes and new forms of contents crop up. I can't stay on top of it myself, so I'm always looking for suggestions/maintainers from anywhere.

If you have a look at the category tree, where do you think video essays would go in that?

It will be here: https://allaboutberlin.com/guides/german-health-insurance

The guide is pretty much finished and already reviewed by 3 competing brokers. If you email me, I'll send it to you. I'm now working on the recommendation tool, then the satellite guides.

TK is great if you're on public health insurance. If you qualify for switching to private health insurance, you might find cheaper and better contracts for the cost of more bureaucracy and that having kids costs extra.

I appreciate that! Honestly, I'm not sure. Part of the reason I felt competent enough to write about Estonia is because I had lived there and worked for the government, so I thought I had a pretty decent insider look into how things worked and could tell the story well and I'm not sure I could replicate that in Poland. Although I do think a comparative study across some digital leaders (ex. Estonia, Poland, Ukraine) or even just comparing the Baltic states on topics like digitalization would be super interesting.

Not right now, working to get it on Bookshop

I am the first to admit that the UI does not make sense to people who have not seen it in use. I have done some user onboarding sessions where you just “let them ask questions and click” and it goes terribly if it’s any consolation, I have never gotten that feedback after showing people how to use it (quite the opposite actually!)

Wrt to the expression, all expressions created today succeeded, so if you’re still seeing a progress bar let me know as that’s a bug. It’s possible something failed with the live updates, or it does take several minutes to create an expression (depending on servers, it can take up to 10 minutes at times, although the typical timing is 2-4 minutes depending on the expression)

If you click on any of the review methodologies, it will start reviewing any of your successful expression. From there, the experience should be a lot more explorer-friendly :)

What it’s doing is: analyzing the sentence, splitting it into phrases, aligning it across all languages, tagging all of the gender/case/tense/etc, researching pronunciation, generating audio, aligning the audio, prioritizing the words (across several axes), and generation explanations/dictionary for each individual word

I made some quick live videos today:

https://x.com/barrelltech/status/1917093849219895715?s=61

Please excuse the video quality, there will be better paced/audio/scripted demos soon!

I would have to know more about your circumstances before I could make genuine recommendations. But as an autodidactal programmer, I can serve as somewhat of an authority on this manner ;) The value I get out of the platform is:

1. I can study all my languages on the same platform. For my, having studied 30+ languages (note: not claiming to speak them), I just want to “do my languages”. I can study dialectal Arabic, minority languages, archaic languages, and the major languages, all in a nice consistent and (if I can say so myself) beautiful UI.

2. Everything is heavily annotated with all the information you could ever need. This means that I add flashcards, and when I’m learning them, I have the gender, cases, tenses, agglutination, phonetics, translations, audio, conjugation/declention tables, character breakdowns, mutations, idioms, multilword expressions, roots, etymology, etc etc (the list really does go on) at my fingertips. This means I just go through my flashcards, and when I have a question, I get an answer. If I have more questions, I have a context aware chat integrated. For me, this is the autodidactal dream come true.

3. Personally, I really love SRS. I also really hate SRS. If I have to study the words “dog”, “walk”, & “morning” - and I have a sentence “I walk my dog in the morning”, I just want to study that one one sentence and be done with it. Also, I really want to just be able to play audio sentences and listen to them while cooking/cleaning/walking my dog. Or do a free recall sessions, write down everything I remember from yesterday, and skip those reviews today (it’s more effective than SRS anyways).

Lastly, WRT to creating your own flashcards: You can still create flashcards manually on Phrasing - I agree the act of creating flashcards is beneficial, I’m not trying to take that away from anyone - but I’m not sure I buy it’s the highest leverage way for one to spend their time. At least for me, it definitely is not. I would rather skip that (admittedly beneficial) step, and move onto the next step. YMMV

It’s really hard to narrow the list down to three, I have a hundred things I want to say, but I’ll leave it here. Due to popular demand, I recorded a few live demos today so you can see it in action:

https://x.com/barrelltech/status/1917093849219895715?s=61

Higher quality demos will come in time!

Let me know your thoughts, I’m happy to dive deeper into any of this (I mean I could talk about Phrasing for literally days on end)

EIDT: s/extinct/archaic

Oh no, the website is brand new, it should be working everywhere. I'll have to dig up an older android, I should have one somewhere.

Languages below, if you know their alpha 3 code. Currently having some issues with Thai and Zulu though, so they're temporarily disabled until I have time to fix them.

I have not ~tested~ verified it for Latvian, I would be curious to hear your thoughts. It has been working pretty well for Maltese, Albanian and Macedonian though, which should be lower resource than Latvian!

As mentioned elsewhere, the first time user experience is abysmal. If you reach out though we can hop on a call and get you set up - or in a few weeks I'll have a video done and up. In the meantime, you should be able to create an expression (in the nav bar for desktop and mobile) fairly intuitively.

afr, amh, ara, ara-are, ara-bhr, ara-dza, ara-egy, ara-irq, ara-jor, ara-kwt, ara-lbn, ara-lby, ara-mar, ara-omn, ara-qat, ara-sau, ara-syr, ara-tun, ara-yem, asm, aze, bel, ben, bos, bul, bxr, cat, ces, chu, cop, cym, dan, deu, ell, eng, est, eus, fao, fas, fil, fin, fra, fro, gla, gle, glg, glv, got, grc, guj, hbo, heb, hin, hrv, hsb, hun, hyw, iku, ind, isl, ita, jav, jpn, kan, kat, kaz, khm, kir, kmr, kor, lao, lat, lav, lij, lit, ltc, lzh, mal, mar, mkd, mlt, mon, msa, mya, myv, nan, nep, nld, nno, nob, ori, orv, pan, pcm, pol, por, por-bra, por-prt, pus, qaf, qpm, ron, rus, san, sin, slk, slv, sme, som, spa, spa-arg, spa-bol, spa-chl, spa-col, spa-cri, spa-cub, spa-dom, spa-ecu, spa-esp, spa-gnq, spa-gtm, spa-hnd, spa-mex, spa-nic, spa-pan, spa-per, spa-pri, spa-pry, spa-slv, spa-ury, spa-usa, spa-ven, sqi, srp, sun, swa, swe, tam, tel, tha, tur, uig, ukr, urd, uzb, vie, wol, wuu, yue, zho, zht, zul

EDIT: I have tested it for Latvian, I know it technically works. I however have not had any Latvian speakers review it's quality

I made some quick live videos today:

https://x.com/barrelltech/status/1917093849219895715?s=61

Please excuse the video quality, there will be better paced/audio/scripted demos soon!

Block rendering errors on the media page is new to me. I will look into it.

The reference language error should not be shown (I mean it’s not incorrect, but there is a “no expressions error” that should take precedence).

A video is coming :) I didn’t expect so much interest from a comment in this thread. If you get in touch, I can walk you through it personally, otherwise check back in a couple weeks and there will be a video overview.

I made some quick live videos today:

https://x.com/barrelltech/status/1917093849219895715?s=61

Please excuse the video quality, there will be better paced/audio/scripted demos soon!

It does really well, 95% of the time. The application was built to jump in at any levels - find a movie you want to watch, align the subtitles, see the most important words, create expressions with words you don't know, and the SRS should focus on the words most important to you.

For my Dutch (which was probably once a high B2, now probably a low B1) I only use the audio review when walking my dog or cooking. It plays the audio of the cards in a playlist, so I practice hearing and repeating them.

It's not so self serve at the moment, but if you get in touch I can get you up and running.

Thanks for the feedback! I agree with you completely.

1. I just started the marketing website a few weeks ago, and if you can believe it, I didn't readily have that information. One of my tasks last week was to compile a list of languages that could work, write some tests for all of the languages, and get a list of supported languages. I have that list now, I just need to put it on the marketing page.

2. As mentioned in other comments, I'm working on a video. I'm preferring to fix glaring issues before making the video, although at this point I'm verrrrrrry close. I have started scripting it, but it takes a lot of time to make a good video (1-2 full days if I don't want to edit it).

Your feedback is completely valid, and they're both reasons why I'm not really marketing the product yet. This thread seemed like a good middle ground though as having some people using all the languages would be really helpful. Also, I've genuinely been loving using it and want to share.

It's just me working on it, so these things are coming, but everything takes a while! Hopefully these didn't sour you on the project permanently :)

EDIT: And yes, it supports Hungarian :)