1. Set up SSH access to your PC (I recommend tailscale)

2. Install and start tmux session on your PC so it stays synced and survives disconnects and what not.

3. Install Termux, SSH into your PC, attach to the tmux session.

Bonus: tmux layout will scale nicely to your phone aspect ratio

You can also setup mosh connection if you expect signal to be frequently lost due to poor network quality, etc.

Don't forget to intensely shake your head after consumption for a proper brain flush

Thoughts on possible implications for users in foreseeable future? We built a lot using dbt and can't really think of going back or switching to alternatives

Fivetran isn't really much of a transformation layer so this is likely just a move to lock-in customers of both companies by upselling an ingestion/transformation layer to existing customers.

The bigger question mark to me is that Fivetran recently acquired Tobiko, the company behind a dbt competitor SQLMesh. The Tobiko team said their focus has been on dbt-compatibility because a lot of Fivetran customers use dbt for their transformation layer. I fear it may have just been a way to get rid of competition leading up to this deal. I can't imagine Fivetran spent a ton of money just to have 2 products that do very similar things.

We use both open-source SQLMesh as well as their cloud offering Tobiko Cloud. Following the acquisition, we were annoyed that focus was going to go to dbt compatibility because there was a bunch of stuff on their roadmap that would help us that was now deprioritized. Thankfully, they still offer great support to us and delivered a few features that have given us some quality of life improvements. With this announcement, I'm worried we're going to end up being forced to migrate to dbt...

Full disclosure: I am a PM at Fivetran who is very excited about this.

We are fully committed to open source dbt and don't want to build a 'walled garden'. Interoperability is one of the key value propositions of both Fivetran and dbt. While I'm biased, I think the main implications for users is that their favorite tooling will be with one vendor who cares about what makes them great.

You can read a bit more here: https://www.fivetran.com/blog/the-era-of-open-data-infrastru...

Thanks!

What's the lock-in?

I'm wondering too. We run dbt on-prem. Worst that could happen is we don't get any more free updates. But we have the software and it will continue to run.

the concern is that dbt-core will become stagnant.

It basically already has since they started developing dbt Fusion, so in that respect this probably doesn’t change much.

I expect they’ll keep developing Fusion but possibly as even more of a commercial-only offering than it already was.

What's the problem specifically? Are you banking on some future features? Can't fix the bugs yourself? Worried it won't be compatible with future data warehouses?

I know people don't like it these days, but you can just continue to run old software.

dbt in particular is effectively useless without maintained and up-to-date connectors to your particular database.

Do your database vendors often do breaking changes to their protocols? dbt just generates SQL and that's not going anywhere.

The cloud component is probably sticky if you have come to rely on those parts.

Are you using their cloud offering or just the software itself?

Just the software

LLMs have brought back my excitement with coding.

At work, they help me to kickstart a task - taking the first step is very often the hardest part. It helps me grok new codebases and write boring parts.

But side projects is where the real fun starts - I can materialize random ideas extremely quickly. No more hours spent on writing boilerplate or fighting the tooling. I can delegate the parts I'm not good at the agent. Or one-prompt a feature, if I don't like the result or it doesn't work, I roll it back.

It's far from sustainable (for now)

Assuming you have to generate new content for each viewer second watched yes it won't pencil out. But if you have a library of tons of content you can keep building out...

It's LinkedIn so I wouldn't be surprised if opting out did nothing

So that's the new default for Rails? Seems like a new way for people to shot themselves in foot, for those who are used to the usual postgres + separate services setup. Especially that it's advertised as simpler and therefore is tempting to try out without considering such edge cases. I mean, it's the usual route (from my experience) for successful Rails apps to scale horizontally

What's new is that it's now being recommended for use in production, not just development. It's also now what they want you to use for caching, artifact storage and just about everything else that use to use "Technology X".

I can see the reasoning behind it. Simplicity. But I'm waiting for the horror stories like this one to die down before making a final judgement. It just doesn't _feel_ right to me after doing it differently for 20+ years.

Sqlite has always been the default database for Rails, even though it probably hasn't been the most common deployment choice.

Kind of but not really. Rails has defaulted to creating a project with an SQLite database in development for awhile now. But the expectation has always been you move to what you need.

SQLite has simply become "good enough" for production. So now you can take that dev default and simply deploy it for apps of a certain size and category if you want.

Well, you can run apps on any less capable device with ssh and proper terminal display. You can limit data usage by offloading video buffering to the host (however not sure if that's net positive saving). And put the host behind VPN to avoid getting region blocks.

I actually used to tunnel Netscape Navigator via SSH to my Commodore Amiga desktop via an Xorg server way back in the 56K phone modem Internet days from my ISP's SSH user account login, since Amiga didn't have Netscape (and even if it did, the Amiga likely would have choked on it, massive and bloated as Netscape was), and the browser AmigaOS did have just wasn't up to the task of normal day-to-day usage of the Web as it existed back then. Fun times.

Sure am glad of the broadband Internet and modern "powerhouse" PCs we have so readily available today. Hell, even the computer most everyone carries in their pocket these days is infinitely more powerful than the average desktop machines of my childhood. :)

Oops, we've invented X

Because you posted a success story about LLM usage on HN

Well, understood, but that part between the lines is not my fault?

Nah, never implied that