lars_francke

Disclaimer: I'm totally biased here.

I talked with a colleague from Oxide in 2024 about your Kubernetes story and he said back then "not yet but soon-ish". Seems like soon-ish is now :)

We said we'd talk again when that happens but he's since left Oxide. If you (or well...your customers) are interested in a Kubernetes native data platform 100% open source we'd be very happy to talk about how that could work easily. As it's "just" Kubernetes it should be trivial but we'd be happy to test and add you to our list: https://docs.stackable.tech/home/stable/kubernetes/#supporte...

The offer stands. If you're interested, my mail is in my HN profile. https://stackable.tech

Partially. Yes. Look at the budgets of these orgs and you'll see what I mean.

I use the term similar to who it's used for non-profit. The orgs I'm involved with are almost exclusively not involved in the actual standards creation.

If the secretariats were to shut down tomorrow I'd say the actual work on the standards could continue without anyone noticing.

There is a reason that at least the EU is considering modernizing the system. https://single-market-economy.ec.europa.eu/consultations/pub...

ISO, CEN, CENELEC, ETSI are stuck very much in the past.

So yes. Overhead.

As someone working in standardization: I don't know any standardization organization where the people doing the actual work of writing standards are paid for their work. I certainly am not.

In the organizations I know - including ISO - the money is basically exclusively spent on "overhead".

Because of bureacratic overhead. We did employee people from other EU countries in the past because I personally honestly don't care where someone is from or where someone lives. But you need to learn the rules of every country. And they are often subtly different. We even struggle with UK & Germany already. It's annoying. And we have like 0,1 FTE to take care of HR stuff and anything "out of the ordinary" eats into that time and money budget.

The rules need to change. The rules will not change. So we're stuck.

It's not us. It's the customers that struggle with the pace already. In the enterprise world we operate in having more than one release a year is already considered breakneck speed.

Thank you :)

Stackable | Product Engineer - Web & UI | Remote (Germany/UK) | ~20 people | Full or part-time

IMPORTANT: ONLY APPLY IF YOU ARE PHYSICALLY BASED IN GERMANY OR UK! (I received almost 100 applications, about 85 of them did not read the details)

We build an open source data platform on Kubernetes. The product (Stackable Data Platform - SDP) bundles 12+ open source data tools (Trino, Apache Kafka, Apache Airflow, Apache Superset, etc.) and makes them work together. Right now everything is Infrastructure-as-Code only. We need someone to build a UI layer on top.

No mockups, no Figma, no product roadmap waiting for you I'm afraid, just a bunch of ideas :) We're all backend people and need someone to own this end to end. Play around with our product stack, figure out what customers actually need, build it.

Stack (because we got started without you...): Svelte/SvelteKit, TypeScript, TailwindCSS. SDP itself is Rust, Java, Python on K8s. Three releases a year. Not SaaS. Quality matters.

100% remote, Germany (UK possible) only, no other countries, sorry. No Scrum, minimal meetings. Interview process is usually just two conversations. The position isn't on our homepage yet as I'm on vacation. Send a mail to recruiting@stackable.tech and I'll get back to you next week. Please mention HN in your mail. My personal mail is also in my HN profile. I'm the Co-founder and CTO.

https://github.com/stackabletech/ | https://stackable.tech

Because that's what everyone is used to.

Maplibre supports different projections if you want.

This is a terrible idea in my opinion and it's been tried/is being tried by services like thanks.dev. Yes, we need something here but this is not it. The reality is more complex.

It doesn't work well in practice. Because then people like https://github.com/sindresorhus?tab=repositories&type=source would get a shit ton of money because of the pure number of dependencies. And yes our stack also contains his code somewhere in a debug UI but our main product is entirely written in a different programming language with way fewer dependencies but if one of them goes away we'd be in trouble. In other words: Dependency count is not a good metric for this.

GitHub actually offers something in that direction: https://github.com/sponsors/explore

My "idea": Lots of companies will have to create SBOMs anyway. Take all of those but also scan your machines and take all the open source software running on there (your package.lock does not contain VLC etc.) and throw it in a big company wide BOM, then somehow prioritise those using algorithms, data and just manual voting and then upload that to some distributor who then distributes this to all the relevant organisations and people and then (crucially) sends me (as a company) an invoice.

We've tried doing the right thing but sponsoring is hard - it works differently for every project/foundation and the administrative overhead is huge.

The reality is that "we" as an open-source community suck at taking money and I believe this is partially on us.

A battery truck is allowed to have 42 tons.

I'm no expert but a far stretch but if this most basic fact is already wrong then my trust in the remaining stuff diminishes. On top of that they is only relevant if all truck loads were limited by weight.

So, I believe that argument to be wrong in its entirety. And if we then factor in the CO2 costs, hydrogen is the clear loser in all regards.

There's a fabulous YouTube channel on electric trucks in Germany if the topic interests you.

German version: https://youtube.com/@elektrotrucker

English: https://youtube.com/@electrictrucker this has fewer and shorter videos unfortunately.

In case it makes you feel better: I wondered the same thing. It's not explained anywhere on the blog post. In that poste they assume everyone knows how pricing works already I guess.

I'm using a different approach for local testing where I don't want to redownload images over and over: https://github.com/stackabletech/k8s-local-dev

Basically it's a k3s configured to use a local mirror and that local mirror is running the Zot registry (https://zotregistry.dev/v2.1.8/). It is configured to automatically expired old images so my local hard drive isn't filled up).

That is not true.

There is a process, it's usually tedious but it exists. I did it for Singapore, the US and Israel. They mostly took multiple months but I never wanted to take any chances. For the US it was a "B-1 in lieu of H-1B" visa for example.

Attending a conference is something different than what these workers did. There are rules around what a "business trip" is and what is not and what "work" is.

Please don't run them at night to protect animals like hedgehogs and others that are active at night.

In which way is the situation far worse in Europe?

Thank you!

I've been using SXN-210 for over a decade and I love them.

Not commenting on the substance but on the https://opensourcemaintenancefee.org/ homepage itself. It only works in dark mode and is unreadable in light mode.

The repo doesn't allow opening issues. Maybe the author reads here... (long shot)

In Europe this is already happening. There are dedicated companies for this (Milence) and the traditional operators have truck charging parks as well.

These do exist already. The Electrictrucker also has a video about this. They can be combined with a EV truck as a range extender or with a regular truck to save on gas/diesel.

You did. I'm sorry, my bad.

He has a german channel which has more and longer content: https://www.youtube.com/@elektrotrucker

He also has an english language version of his channel: https://www.youtube.com/@electrictrucker

It's very worth watching.

I think he's proven that single driver long haul freight in Western Europe (which seems to be a good chunk of truck trips) is perfectly doable. Just two weeks ago he did a 4.500km trip around Europe which is about the maximum you can do given the law on driving times.

The same is then true for the shorter trips (round trips etc.)

And the European Commission has just decided that electric trucks don't have to pay road toll until at least 2031.

Better for the environment, more comfortable, quieter, cheaper in the long run, ... what's not to like.

And yes: There are still some use-cases where non-EV trucks are "better" by some metrics but that's definitely not the case anymore everywhere.

This particular case is US only.

The rest of the world might decide differently.

Agreed. And for me that has nothing to do with announcement vs. notes.

"We sped up new file creation, so you can now create a new file in under 20ms, a 100x speedup from v1.2."

No. You had a big bug. You fixed it. Excellent. When I'm affected by the bug I'd search for "deadlock" or "bug " in the announcement and this is hiding it behind marketing speak. I don't think that's honest. So for me the perfect announcements would be the red ones.

Our customers are mostly on-prem and we usually hope they have a storage appliance (Netapp, Dell, VAST, ...).

But for those wanting other options and self-host something it's not clear cut.

Which alternatives are there though? Honest question.

- Ceph

- SeaweedFS

- Garage

- CubeFS

- Apache Ozone

Do any of them fill the niche that MinIO sits in? (Relatively easy to set up and operate and with all the typical enterprise bells & whistles)

Ozone/Ceph -> Way more complicated

> I wish the native backend supported pushing/pulling changes.

0.29.0 is working towards support here. It's only the first step as far as I understand (not following too closely): https://github.com/jj-vcs/jj/releases/tag/v0.29.0 git.write-change-id-header

> If you forget to start a new rev before you (or your LLM) touch the repo, it's a little bit of a pain to go back and split the changes into a new rev.

This is my biggest pain point at the moment, yes. I think it could be partially solved by better documentation already. I contributed a bit in the past but haven't had the time to look at this bit yet.

This is at least partially subjective.

https://news.ycombinator.com/item?id=32148007

https://news.ycombinator.com/item?id=35299665

Ceph would be a theoretical option, but a) we don't have a lot of experience with it and b) it's relatively complex to operate. We'd really love to add a lighter option to our stack that's under the stewardship of a foundation.

Anyone looking for an open-source Cloudera alternative based on Kubernetes operators. We're building one (~5 years old now): https://stackable.tech/ & https://github.com/stackabletech/

On-premise open-source S3 is a problem though. MinIO is not something we're touching and other than that it looks a bit empty with enterprise ready solutions.