I like using the em dash, but I try to limit it now—because chatgpt loves it even more. Maybe it's for the better, some people seem to have strong opinions about it.

I find it distracting, I think because I'm so used to humans using two hyphens as a substitute, sometimes with spaces around them, that a real em-dash used what I suppose is correctly looks too small and tight.

Yeah, understandable.

I’ve actually never been a fan of the “correct usage” that forgoes any spacing around it. I find it crowded and jarring.

I like to use a “thin space” around an em dash — it breaks it up without adding too much space.

As usual, isn't the next question... "What are you trying to build?"

I build most projects in Elixir/Phoenix these days, but wouldn't flinch at ROR if you are comfortable with it.

Right now, an AI tool that generates mockups for branding agencies. But I’m still validating the idea so who knows. Ideally, I would like a stack that would work for most SaaS I may think of building. Tempted to give Elixir Phoenix a try. I briefly tried it a few years back and it just felt right.

As it's just you I'd stick with Ruby on Rails 8[1] as you already know it and I think it could realistically easily achieve what you're proposing.

There's lots of libraries for calling out external AI services. e.g. something like FastMCP[2]. From the sound of it that's all you need.

I'd use Hotwire[3] for the frontend and Hotwire Native if you want to rollout an app version quickly. I'd back it with SolidCache, SolidQueue, etc

I'd use Kamal[4] to run it on cheap hosting using on something cheap from Hetzner.

1. https://rubyonrails.org/

2. https://github.com/yjacquin/fast-mcp

3. https://hotwired.dev/

4. https://kamal-deploy.org/

>What are you trying to build?

Indeed. Not everything even has to be a "web app" in the first place.

I love this, I've primarily been working in Elixir for a few years now and this is neat to see!

Yeah, unfortunately this is true. A street near my house has a limit of 40mph and people would regularly drive 60 mph+, sometimes someone would pass me doing 65+mph (it's a no-passing residential road).

Eventually someone died, and they added a lot of traffic-calming changes to the road. It's much nicer now, but a shame that someone had to die to change it.

I'm working on a solution for gathering product metrics and making sure applications keep running — when you don't want to install or maintain a lot of extra stuff. https://flexlogs.com

... also continuing to not add features to my (not-much-of-a) system for getting more done each week. https://carpeweekem.com

I've also been cautiously adding AI-powered features to my Heroku autoscaling tool (https://flightformation.com/) and a simple free-text time/date input has been the most popular (demo: https://x.com/ejschmitt/status/1893268742760448497)

Glad I'm not the only one that feels this way. For a small application when you just want some metrics and observability, it's a big burden to get it all working.

On my own projects, I send the metrics I care about out through the logs and have another project I run collect and aggregate them from the logs. Probably “wrong” but it works and it's easy to set up.

I usually implement the whole username/password auth flow, but recently used only magic links for a simple application.

Since the application only sends a weekly email (a markdown template for goal/task tracking) it seemed easier to just use a magic link, only.

I am happy at how much easier the auth code ended up, and fail to see much downside for such an application.

I'm not sure it would be a good system for more complex apps and services.

I have a system where users log in extremely infrequently. Tempting to move away from username & password because they just reset every time anyway.

Yeah - that's my thought too, the service I use them for is not something people often log into. Sometimes never.

I love my sawstop, it's a great machine.

Often times when a product has some patent-protected feature, the product itself is substandard, but I have not found that case with sawstop. It's one of my highest quality tools.

It would be nice if the mechanism wasn't so destructive. I accidentally had an aluminum fence just a fraction of a mm too close, and it touched the blade. I was using a dado stack, and it did a number on the carbide teeth of the blades. Good dado sets are not cheap, nor is the sawstop cartridge.

We had a saw with a mechanism that was non destructive and SawStop sued them out of the country (Bosch REAXX). It's why my feelings on SawStop are complicated, they say they're all about safety and willing to work with others but stomped out the only one that tried.

Bosch REAX used compressed air cylinders to drop the blade without damaging it. They got sued into oblivion by SawStop because SawStop was somehow granted a patent on the idea of stopping a blade quickly. As a side fact, the Bosch sensor electronics weren't done properly and could sometimes be affected by BlueTooth.

The whole "releasing our patent" is simply SawStop's way of trying to lock out the competition. All their competitors (including Bosch) have said that it will take several years before they could develop an alternative product leaving them in violation.

Finally, the regulation SawStop is trying to force doesn't even solve the injury problems for a few reasons.

The biggest is that CPSC does NOT affect commercial saws. As it turns out, hobbyists don't have as many injuries as you might think because they don't use their saws all the time and they have a very healthy respect for them (there are exceptions of course). Most serious injuries happen because the guy at the commercial shop has become too complacent and made a mistake after a long day at work. This ruling does nothing to change that situation.

You also can't fix stupid. If blade guards and riving knives are left on saws, the chances of injury are incredibly low, but people choose to remove one or both of these. They'll also turn off the safety features and do something they shouldn't. SawStop safety is over-represented because the people who spend the extra money for one are already predisposed to take safety seriously.

This leads to the price issue. Table saw prices will go up from $220 up to a minimum of $600 or more. This increases the risk of someone not having that much money and then turning their circular saw upside down making an incredibly dangerous table saw without a blade guard, riving knife, or even a parallel fence massively increasing the baseline risk for injury.

I love the idea of SawStop and I think it's an amazing safety device, but after reading the arguments on all sides, I think we should leave the current saws situation alone and instead simply require each saw manufacturer to offer at least one AIM model in their product lineup by 2032 or so (while maybe getting the courts to fix up the colossal screwups they made with the SawStop patents). This will give them time to develop alternatives and maybe drive down prices over time until it finally (hopefully) makes economic sense to only sell AIM devices.

I'm struggling to imagine a scenario where dadoing <1mm from the fence was a good idea in the first place. I'm assuming you're talking about a miter/crosscut fence but still...

I've stopped myself just in time from triggering my Sawstop while using an aluminum miter gauge.

The scenario is that the gauge was set with reasonable (1cm) clearance at one angle, and then I changed the angle without re-checking the clearance, and the back side of the fence swiveled into the blade path. Pythagorean fail.

Or, perhaps with GP comment, used a setup with reasonable clearance for a regular blade, and then put in a wider dado, which ate into the clearance.

Cutting a rabbet? Probably should have used a sacrificial fence, but it's easy to see how this could happen.

I had a sacrificial fence (wood) on the aluminum fence to reduce blowout. I had the aluminum extend to near the blade to reduce the deflection of the whole thing. It was too close.

Really, I should have sacrificed a crosscut sled for dado usage, but it really chews them up.

Could be an aluminum extrusion or something in a miter gauge or cross-cut sled.

+1, I've had mine for 5+ years and it is still genuinely a joy to use. I went with the "buy your last tool first" approach and splurged on a 5HP ICS and don't regret a single penny spent on it.

This is great.

I used to avoid burl when I would make wood rings [1]. It's just very hard to predict how it'll behave if you try to steam bend it.

[1] https://woodaround.com

why'd you stop?

Took too much time

Those are beautiful! And I love the way you get seemingly continuous grain all the way around. If you no longer work on these, I would love to learn more about the processes you use for the pressure infusing and the finishing.

Maybe prefill it with an example?

It's hard to know at first glance if it's real or not. Many of these things end up abandon on the web, so it's hard to know the real ones that still work.

This was a very good suggestion. Updated.

Thanks..