Aaron Francis

The team is back from Christmas break, with a lot to talk about.

Learn more →

Transcript

Aaron

00:00:00 – 00:00:16

K. We got a lot probably to cover having been gone for, like, 3 weeks. So it's good to see y'all again. Colleen, I was just hearing about your road trip from hell, so I'm glad you made it back.

Colleen

00:00:16 – 00:00:18

Thanks. Me too.

Aaron

00:00:21 – 00:00:22

Sean, how was your break?

Sean

00:00:24 – 00:00:45

Yeah. It was good. So I got a I got an update. So I know that Colleen's kids listen to this podcast, so I I always have these updates where I'm like, what would Colleen's kids like to hear about? So, you know, whenever I'm talking about, like, almost peeing my pants at the airport or whatever, like, we know who the audience is.

Sean

00:00:47 – 00:01:00

So, anyway, my update for this week is I unlocked the Apex Predator tag in Battlefield 2042. This means I got 42 kills in 1 round against a bunch of teenagers that are way better at this game than me.

Aaron

00:01:01 – 00:01:05

That, I have no context for that, but it sounds like a big deal.

Colleen

00:01:07 – 00:01:08

Sounds good.

Aaron

00:01:09 – 00:01:13

I don't even know what did you say? Battlefield 2042?

Sean

00:01:14 – 00:01:18

Yeah. Yeah. It's just like a 1st person shooter thing.

Aaron

00:01:20 – 00:01:20

Yeah. A couple

Sean

00:01:20 – 00:01:38

of years ago, I was like, I have always been terrible at first person shooters, always terrible. And I realized this is because I never put any time into 1. So I was like, I needed desperately a hobby that I could take way too seriously to, like, blow off that energy.

Aaron

00:01:38 – 00:01:39

Mhmm. I

Sean

00:01:39 – 00:01:51

have that energy. Right? Like, I need to blow that off so that way I'm not, you know, bringing that everywhere else I go. So I'm like, okay, what's it gonna be? So I picked at that time, it was the Star Wars 1.

Sean

00:01:51 – 00:02:01

It was like Battlefront 2. So I just really got into that. I completely sucked. It was ridiculous. I couldn't even get, like, one elimination per round.

Sean

00:02:01 – 00:02:41

And then after, like, 2 years of that being, like, my my, like, game that I played when I would go play, I was able to get to where I'm, like, most of the time I was number 1 or number 2 out of the 40 people playing the match. And and so then Battlefield is a more serious game. And so that game, again, I was kinda like midfield when I first started, and then I kinda figured out how it goes and then decided to see what I could do. So then, yeah, I got this this Apex Predator tag. So now when I when I sign up my little player card, it says Apex Predator on it.

Sean

00:02:41 – 00:02:43

So everybody knows I'm, like, real good.

Aaron

00:02:44 – 00:02:48

So all the 12 year olds are super intimidated by you? Yeah. Yeah. Does your

Sean

00:02:48 – 00:02:57

son play? Yeah. So he plays that game he doesn't play. That one's just, like, too directly violent, if that makes sense. Battlefront 2.

Sean

00:02:57 – 00:03:08

Yeah. He used to play that with me. That was, like, one of the first games he ever played, actually. Because I was playing it all the time, and he would come down, and I would be like, it's Star Wars. I'm just, like, blowing up robots, like, whatever.

Sean

00:03:08 – 00:03:19

He watches Star Wars. So we would do that, and he would try and play it. And he's, he still likes that game a lot. He could be way better than me at this point. Like like, he prop like, give him a year.

Sean

00:03:19 – 00:03:28

If I if I let him play, you know, battlefield or whatever, like, as much as he would like, she would be better than me in, like, a year at this point. I'm sure.

Aaron

00:03:31 – 00:03:38

Yeah. Well, that's exactly the kind of thing you hope to do over Christmas break. So that sounds like a big win.

Sean

00:03:39 – 00:03:45

Yes. I achieved one blow off thing. So that was nice. And then you

Aaron

00:03:45 – 00:03:51

achieved a bunch of you you achieved a bunch of other stuff too, though. You, like, did a 1,000,000 things around the house, didn't you?

Sean

00:03:51 – 00:04:14

Yeah. Well, okay. So, I mean, so actually, there's a ton of work stuff that I got done. So there was, like so I ended my week at work with a very important time sensitive project that had to be finished. So I spent, like, Wednesday, Thursday, Friday working 15 hour days, and then went into that weekend where we were getting set up for our customer.

Aaron

00:04:14 – 00:04:15

Oh, yeah. We Are

Sean

00:04:15 – 00:04:17

we allowed to say his his name?

Aaron

00:04:17 – 00:04:17

Yeah. Are you

Sean

00:04:17 – 00:04:17

okay with that?

Aaron

00:04:17 – 00:04:18

Neil.

Sean

00:04:18 – 00:04:39

Yeah. So we're getting set up for Neil on Monday, and I'm like, okay. We gotta get this build thing working. So then that turned out to just just be a nightmare, for a million reasons. But in the end, like, again, so just stacking on the 15 hour days on Saturday and Sunday and then on Monday again.

Sean

00:04:40 – 00:04:59

And then, finally, in the last hour, was it Monday night? We we discovered or I discovered that it's, like, gonna be impossible for me to use the, like, built in Vue class setup. Like, because that's my shortcut for this stuff is you choose the library. Like, in React, it's the same thing. There's create React, app, and then there's and it's you choose those things.

Sean

00:04:59 – 00:05:28

That way, you don't have to understand how to make a bundle for NPM or whatever, and then you just run their script. And, like, at the last hour, I realized, like, it's never gonna work because it doesn't support the type of module system that we need to have. So because it's view 2 and view 2 and view 3 are this whole in between state. And it's just like so, like, a lot of the view 2 stuff is not getting updated. Anyway, so I'm, like, rip it out, replace the whole thing with roll up, got it done at, like, midnight.

Sean

00:05:28 – 00:05:35

Aaron and I are both working on this together. We got it done. We're ready to go for Tuesday morning. And then and then our buddy Neil rescheduled, which

Aaron

00:05:37 – 00:05:40

Yeah. So Neil Neil DM'd me Tuesday morning,

Sean

00:05:40 – 00:05:40

and I'm

Aaron

00:05:40 – 00:05:44

like, hey, bad news. I'm sick. Can we push it? And I was like

Colleen

00:05:44 – 00:05:45

Oh my goodness.

Aaron

00:05:46 – 00:05:47

Yeah. We can.

Sean

00:05:49 – 00:05:51

It's totally it's totally fine.

Aaron

00:05:51 – 00:05:53

Yeah. It was nice. He was like, I'm not anchoring

Sean

00:05:53 – 00:05:54

him at all.

Aaron

00:05:54 – 00:06:16

Can we push it or does that mess up your timeline? Man, let me tell you about our timeline. So he is super nice about it. But we got him rescheduled. So, Thursday morning, we're gonna onboard client number 1 again.

Colleen

00:06:17 – 00:06:18

Thursday this week.

Aaron

00:06:18 – 00:06:19

This week. Nice. Yes.

Colleen

00:06:20 – 00:06:21

That's awesome. That's exciting.

Aaron

00:06:22 – 00:06:47

I actually did a bunch of work in his, he gave me access to his repo ahead of time. And so I did a bunch of work leading up to it, getting it installed and configured, which is how I ran into a bunch of these issues. So I got a bunch of the low hanging fruit issues out of the way. So those were the things that Sean and I were working on, the ones that were just like, hey. This doesn't even load the page.

Aaron

00:06:48 – 00:06:57

We need to fix this. And now we can run into the issues that are actually app issues and not, you know, build or bundle or whatever. So Right.

Sean

00:06:57 – 00:07:08

Yeah. And I also spent some time with, because honestly, I left that whole situation feeling not confident in, like, my skill set around build doing the bundle. So I was like, oh my god. Alright. This is ridiculous.

Sean

00:07:08 – 00:07:29

Like, we were in the weeds, like, with this stuff. Like, we were deep in the weeds, and which was frustrating because, again, it's like, why on earth do I need to know anything about this? You know? But so I'm in the weeds, and I ended up having to do my own custom, you know, roll up thing. And, you know, I got it to work real fast.

Sean

00:07:29 – 00:08:06

So I kinda was like, alright. I think I've figured out how this stuff works ish. And then but I still wasn't feeling very confident, so I hired a guy who has a whole bunch of Vue packages out there to, like, just go through it with me. And in conclusion, what the pain that I went through is essentially the pain that every package distributor goes through. And you just, like, Google and copy paste from Stack Overflow and, like, continuously, like, read through docs and, like, make tweaks until you get it to work, and then you never touch it again.

Sean

00:08:07 – 00:08:24

And that's basically how it goes even for somebody with I don't know. He's got, like, a dozen different packages out there. So there's probably somebody who's, like, bothered to become an expert in, you know, building bundles or whatever. That's what he was saying. He's like, probably there is somebody, but, like, yeah, dude.

Sean

00:08:24 – 00:08:30

Like, you're you're doing what everybody else is doing. So we are at industry best practices standards is what I'm saying.

Aaron

00:08:32 – 00:08:54

It's a nightmare. It it I remember trying to do this for, I think, the Torchlight JavaScript client or something, And it's just a nightmare. There's there are no definitive answers. And if there are definitive answers, they're 6 months old, which means they're 10 years old. So it's just nothing works.

Aaron

00:08:54 – 00:08:55

It's really frustrating.

Sean

00:08:58 – 00:09:18

Yeah. And then it's it's also like I I need to also have making our thing more complicated is it's like I I need to be able to run it in development. So it's not just a library. I'm not just building a a library. There's all the other kind of stuff that comes along with it, which is what we lost when we ejected from VueClient, which also, by the way, they do not have a way for you to just eject from it.

Sean

00:09:19 – 00:09:40

There's all this there's all these threads where they're like, nah. You shouldn't ever need to eject from Vue, Clyde. Like, we got all the ways that you can, like, override everything. You should never have to if you're ejecting, you're doing something wrong. I'm like, and here we are, and we're in we're it's it's a bit it's a bit frustrating that that's the attitude because it's like, here we are.

Sean

00:09:40 – 00:10:11

Like, we're moving from view 2 to view 3, and, like, things are broken in this, you know, view 2 client package that they offer, and I can't eject. Like, I can't just get the webpack that they have. So, I mean, I because that's what you would want. If you basically, the way I eject was I, like, install uninstall the packages and, like, deleted the config files that they that I'd set up individually and then had to completely redo the entire thing from scratch with with roll up. Yeah.

Colleen

00:10:12 – 00:10:13

Sounds rough.

Sean

00:10:14 – 00:10:28

Yeah. It was ridiculous. But, hey, I have this whole new skill. I'm totally gonna apply that work too because, you know, we need we need a little bit of love over there on our distribution or, like, our package size. So yeah.

Colleen

00:10:31 – 00:10:40

Yeah. So what's the status now? So, I mean, you guys crushed it. Right? You're ready to

Sean

00:10:40 – 00:10:52

It works. It's integrated. Yeah. We can, you know, make changes and deploy stuff. Like, we kinda have a little bit of a janky deploy process right now.

Sean

00:10:52 – 00:11:17

I did not get around with Jeff, the guy I was working with to doing the semantic versioning thing. There's a whole system where you can, like, set up deploys. So that way in your comment, you basically annotate it with the info for the deploy, and then you push it, and then GitHub actions just do everything for you. So we'll we'll wanna do that. But for now, we just manually add a tag and then run the GitHub action.

Sean

00:11:17 – 00:11:36

It's not not really a big deal. But, yeah, as far as I as far as I'm aware, like, right now, everything is integrated, and it's like a functioning product ready to go for Neil. And we'll be ready to make changes, like, as soon as they come up and start actually, like, working with the customer. Just

Colleen

00:11:37 – 00:11:39

Awesome. Yeah. That's super exciting.

Aaron

00:11:40 – 00:11:50

Yeah. Yeah. It's gonna be great. So we'll meet with him on Thursday morning before. He's in the UK, fortunately.

Aaron

00:11:50 – 00:12:22

So we can meet Sean and I will meet with him at 7 before our regular jobs, but, like, right in the middle of his workday. So that'll work perfect. So, yeah, we'll hopefully get him all integrated. And then I have a meeting that same day, actually, over lunch. I have a meeting with a a guy who owns a Laravel agency, and he DM'd me on Twitter and was like, hey, can I see a demo of Refine?

Aaron

00:12:23 – 00:12:41

And I said, yes. You can. I will figure out how to do that. So I'm gonna give him a demo on Thursday, and hopefully I don't I mean, he may have an immediate need for it, but either way, he's the head of an agency. So he'll have a need for it at some point.

Colleen

00:12:41 – 00:12:42

Right.

Aaron

00:12:43 – 00:12:46

So that'll be good. So Thursday is a big day for us.

Colleen

00:12:46 – 00:12:48

Awesome. That's super

Aaron

00:12:50 – 00:13:15

exciting. And I spent some time I spent some time over the break working on our Nova integration. And, Nova is a, admin panel from Laravel. It's the first party one. So I spent some time over the break working on that, and I've got it to a place where it's almost it's almost ready to release.

Aaron

00:13:15 – 00:13:35

I mean and I was telling Sean over the break, we should have started. This is all hindsight bias. We should have started with Nova because it's so constrained. It's like it's the same. Nova is the same on every Nova installation, more or less.

Aaron

00:13:35 – 00:13:59

They don't even support mobile. Like, Nova, if you're using Nova, it's an admin panel. There's no mobile styles for it. It's very constrained. It's not, you know, with with the one with the Vue package that we're doing, we don't really know what people are gonna need from it in terms of styles and setup and everything.

Colleen

00:13:59 – 00:14:00

Right.

Aaron

00:14:01 – 00:14:30

But as I was working with Nova, getting it integrated into Nova, I was realizing how many things I didn't have to think about. Like, I all I have to do is match the styles of Nova, and then I match all the Nova apps that are out there. And so it's been really great. So hopefully hopefully, I'll get it to the point in the next, I don't know, week or 2 where we can I can reach out to some Nova people, and see if they wanna give it a a beta

Sean

00:14:31 – 00:14:50

try? So we had a a blocker on that, which was the way Nova works. It ends up importing 2 versions of Vue Mhmm. With the way we use our package, which by the way, the way I mean, the way we're using the packages and our Vue and our package is totally normal and fine. It is the way that front end packages are distributed with Nova.

Sean

00:14:50 – 00:15:14

That is a little bit weird. Mhmm. So we have we have to do some work to get around that. So I've with Jeff, we came up with a way to get around it, and we should pull this into our budgeting conversation after the call. But we have there's prob there's a way for us to actually use Vue.

Sean

00:15:14 – 00:15:27

So here's the deal. Like, right now, we could remove our importing our calls to import Vue from the package. Aaron already discovered this. If he removes it altogether, it'll work. So we can remove it because there are, I'm using the composition API.

Sean

00:15:27 – 00:15:54

And so because I'm using the composition API, there are equivalents to the functions that I'm using in that package. So we can replace them with those. The issue is as soon as view 3 comes out, then we'll have to fix the view multiple view problem then. So that'll come up. And then also as soon as view 3 comes out, like, Nova will work differently.

Sean

00:15:54 – 00:16:11

So we'll have this whole other thing to figure out probably. So I think that we could come up with a more robust solution. Like, Jeff and I kinda came up with one. It was mostly his idea. And I think we wanna do that.

Sean

00:16:11 – 00:16:20

But in the meantime, I think you should just rip it out, man. Like, you already did it If you wanna just rip it out that way, because I think that's the blocker on Nova right now. Right?

Aaron

00:16:20 – 00:16:45

Yeah. It is, and it is not because I already ripped it out. So there's so what I did was I basically ejected from our package. I took you know, I went to GitHub and copied everything and then deleted all the renderless stuff, all the other stuff. So I just got down just to the components that I needed, and then replaced those couple of calls to import view.

Aaron

00:16:47 – 00:17:25

And then just, like, the date picker, for example, I just went in and edited the date picker to use Nova's, default date picker. So it is right now, and I don't think this is the solution in the long term, but it is right now a completely standalone package independent from our view 2 package. Just so happens, it's mostly a copy paste of our view 2 package. But at this point, it is completely independent, shipped independently. This and this is one of the ways that Nova is great is we don't actually have to publish an NPM package for Nova.

Aaron

00:17:26 – 00:17:47

All we do is we publish the PHP package, and it'll bring down all of the, front end assets with it. So it's kind of we get to sidestep that problem just a little bit. So, yeah, I don't think this is right long term because we're duplicating a lot of stuff, but it's so easy for now.

Sean

00:17:47 – 00:17:57

What you described as, oh, good. We get to work around this because it doesn't have to be built for NPM is the reason why we have the problem to begin with with the multiple views.

Aaron

00:17:58 – 00:18:17

So it's still yeah. It still needs to be it still gets web packed. And so if I were to just if I were to leave the old import view statements in there, we'd end up with the same issue, because the Yes. The resource itself gets web packed. So Yes.

Aaron

00:18:17 – 00:18:48

I still have to work around the issue, but I did and it's fixed and it's it works. Yeah. So, hopefully, I'll finish that, and then I'll have you and Jeff. I mean, all 3 of us can take a look at it and see if, you know, see if there's anything we wanna bring over from the Nova version or change or just ship it and deal with it later, which is kinda what I'm thinking, but we can decide that later.

Sean

00:18:49 – 00:18:54

Yeah. And Jeff's perfect for that because he's both Lervo and Vue person. Exactly.

Aaron

00:18:54 – 00:18:56

Yeah. And he's done some Nova stuff.

Sean

00:18:56 – 00:19:02

So And Nova install. Although he hates working with the front end stuff in Nova. So

Aaron

00:19:02 – 00:19:02

can't It's,

Sean

00:19:03 – 00:19:04

can't blame him.

Aaron

00:19:05 – 00:19:32

It. So it is very, streamlined in that there are a couple ways to do things and only a couple ways to do things. So if you're if you're right down the middle of what they expect you to be doing, I feel like it's really easy. And if not, it's pretty hard, which is, I mean, you kind of expect that with an admin panel. Right.

Aaron

00:19:32 – 00:19:40

Like, you can't do everything, but the things that make sense for their business case, you can do really easily, if that makes sense.

Sean

00:19:40 – 00:20:04

It does make sense. Yeah. So then the other thing, I'm so excited that we're gonna have customers now because, like, we've got a list of stuff that we could work on, but absolutely no data to tell us how or what we should work on and why and what's our priorities. So that's very exciting to me that we'll have some actual input. Like, for example, theming.

Sean

00:20:04 – 00:20:14

Should we do that? I I know exactly what I'm gonna do, and we're gonna do it eventually, for sure. But, like, should that be first? I don't know. I got, like, a kind of a punch list of stuff like that where it's like, should we do it?

Sean

00:20:14 – 00:21:10

And then that's also real important too because I'm realizing that more and more now we have to coordinate between the 3 of us more and more because as we make as we decide to make changes on our end to either Vue or, the Vue front end or the back end, that it's gonna impact Colleen on either the Hotwire side or the rail side. So we have to, you know, sort that out. And, you know, then there's also other things like if we allow people to do a theming system, then it'd be nice if it was kinda the same in, like, how we do it in, like, the React and Vue and Hotwire and Livewire versions. So it adds complexity, you know Like, it it it at least adds complexity in the thought process behind, like, how do we design new features so that that makes sense in all the different places that we're doing it. Or for it can, like, have different features elsewhere.

Sean

00:21:11 – 00:21:21

Either way, it just adds complexity. So we gotta I've been feeling that, like, we need to sync up more on, especially as we have customers and get that feedback loop.

Colleen

00:21:22 – 00:21:44

Yeah. Well, even I I changed the blueprint to allow for these saved filters to go through. So even stuff like that, like, we should have continuity between each back end and front end so we can hook them together in any configuration we want. So what I changed shouldn't impact anything that you guys are doing, but I don't know that I communicated to you guys that I was gonna do

Sean

00:21:44 – 00:21:48

that. Right. We have we have integration risk, like, across all the platforms.

Colleen

00:21:48 – 00:22:00

Exactly. Exactly. And I'm changing a lot because of these kind of, like, I don't wanna say special requirements, but more in-depth requirements. So I think you're absolutely right. Like, it's important that we all stay in sync on this stuff.

Sean

00:22:04 – 00:22:25

Yeah. And then decide. Also, like, maybe the way we do it is we have, like, the official canonical. You know, maybe it's time to get the code out into a gem, and then that one stays in sync with us and then could fork it for the client, and they can have their own or they could be have it completely baked in to bullet train or whatever they wanna do. You know?

Sean

00:22:25 – 00:22:43

Maybe it's, like, time to just, like, split it, and they just they own that IP. It's custom completely custom thing, they probably wanna just do that, I bet. I just pay us forever to customize it.

Colleen

00:22:44 – 00:23:06

Yeah. I don't think we're I still have yeah. I think that's a good a good plan. I don't think we're ready to do that yet on the rails side, even though I did pull that one section out for Jesse. So we need to think about that, actually, because we're gonna start diverging in terms of what he's working on.

Colleen

00:23:06 – 00:23:18

And I need to check back in with him now that the holidays are over to see how that's going. But but I'm we're gonna start diverging from what that is. So we do need a ground source of truth that we can reference. I think that makes sense.

Sean

00:23:18 – 00:23:24

Yes. And I also want you and me to start working towards getting actual rails license customers.

Colleen

00:23:26 – 00:23:29

I have people who are randomly asking me for it. So we need to

Aaron

00:23:31 – 00:23:31

the

Colleen

00:23:31 – 00:23:39

problem it it's good. I I think I am concerned of now, you know, drinking the Aaron Kool Aid here, and I'm concerned

Aaron

00:23:39 – 00:23:42

I don't know. Maybe this is bad. I take back this. I take back my celebration.

Sean

00:23:43 – 00:23:44

Colleen, it's good enough.

Colleen

00:23:44 – 00:23:49

That so but I'm now concerned that we are staking our reputation on having

Sean

00:23:49 – 00:23:52

That's too much anxiety over our reputation.

Colleen

00:23:53 – 00:23:54

Well I

Aaron

00:23:54 – 00:23:58

feel I feel that in my bones, Colleen. I support you. I think

Colleen

00:23:58 – 00:24:04

that is what's setting us apart from everyone throwing their MVP out there. Right? So I do have a couple people

Sean

00:24:04 – 00:24:05

I think that's imaginary.

Colleen

00:24:06 – 00:24:15

It might be. You you might be right. I wanna let's, yeah, I wanna see how it's going with Jesse because they had some really unique,

Aaron

00:24:16 – 00:24:17

Yeah. They did.

Colleen

00:24:17 – 00:24:36

Concern. Like, that concerns isn't the right word. They have a really unique setup. So they have to do a lot of customizing, and I need to check-in with them to see if the information I gave them is enough to move forward there. I really think we need to get I just the hot wire.

Colleen

00:24:36 – 00:24:50

I think that the way it's integrated now with client, we need to pull out the front end and the back end. Right? I just pulled out the back end and gave it to them because they have their own front end, but it needs to have both, I think.

Sean

00:24:50 – 00:24:55

Mhmm. Mhmm. So it'll just ship like a gem? Yeah.

Colleen

00:24:55 – 00:25:12

I think it needs to it needs to just ship because I think it's too confusing right now. When you just give them a back end, like, that is not enough information. I think there's a huge curve there with like, oh, just send us this blueprint. I mean, it's I don't think it's that easy. I think people are gonna want the whole thing.

Colleen

00:25:13 – 00:25:52

And the rail side, I still have the one big so made a lot of progress right before the break on getting those nested filters into the UI and into the SQL. But and the next the next big step is is making these filters have to be self contained, which we have talked about from the beginning. But But sending in the tenancy is really causing problems, when you try to get more complicated. So this tenancy thing is still an issue, and that is my next hurdle to try and sort out is how we're gonna do that. And I don't think we can ship until we have that figured out.

Colleen

00:25:52 – 00:25:54

Yep. So so I agree

Sean

00:25:54 – 00:25:54

with that.

Colleen

00:25:54 – 00:26:03

Sending it in is working now, but it feels like a band aid, and it's really become obvious with nested filters. Because if the filter's not self contained

Aaron

00:26:03 – 00:26:04

Wow.

Colleen

00:26:04 – 00:26:10

You are either sending the tenancy down from the parent filter that's instantiating the child filter, which is

Aaron

00:26:12 – 00:26:13

or I can think about that.

Colleen

00:26:13 – 00:26:33

Yeah. It's like a new problem. I just you know, I spent some time talking to Andrew about it this morning, and that's gotta be solved. So, once that's solved, I think I feel like I said this already, but I will continue to say this. Once that's solved, I think, that'll that'll really make a huge difference.

Colleen

00:26:33 – 00:26:34

And

Aaron

00:26:34 – 00:26:47

Question. Yeah. Is that tenancy thing gonna be an issue for fictional client number 2? Like, is that a common is that a common thing?

Colleen

00:26:47 – 00:27:09

I mean, my I've worked on a lot of rails apps, and it happened. I mean, yeah, it's a thing, but I don't know that it's this common. I mean, the way bullet train and the way so in bullet train, you can reference every single model to a team. Mhmm. And the way the access tenant gem and the access tenant gem is like a big thing a lot of rails develop.

Colleen

00:27:09 – 00:27:16

If you want tenancy, you usually use this access tenant gem. It gives, I guess, a tenant ID on every single model.

Aaron

00:27:16 – 00:27:16

Mhmm.

Colleen

00:27:16 – 00:27:24

And so I don't know that it's that I mean, I don't know. I don't know that it's that common. I mean, we will run into it.

Aaron

00:27:24 – 00:27:31

Question is, do we need to fix that before we find another Oh. Rails client?

Colleen

00:27:31 – 00:27:34

Yeah. Because we need to figure out this this all ties into the initial query problem.

Aaron

00:27:34 – 00:27:44

I do. And now are you requiring that they pass in an initial query versus provide a builder that gives an initial query?

Colleen

00:27:44 – 00:27:45

Correct.

Aaron

00:27:45 – 00:27:46

Got it.

Sean

00:27:46 – 00:28:13

So so but, Colleen, does that just mean that initial handful of clients will have to jump through some hoops to do this initial query thing because it's not it's not like the API that we so desire to eventually present to them. And then within a couple months, we'll be, like, okay. Here's the new thing. We will hand hold your hands and update this new API with you so that it's all good. And it'll be, like, 5 clients and not.

Colleen

00:28:13 – 00:28:37

Yeah. We can totally do that. And that's the way to do it. So I think the way to do it is you say, we if we ship what we have right now to Rails, first of all, has we have to get the Hotwire stuff pulled out and I started that, Sean, so you and I can work on the rest of that together. What we would do is we would say you can't add a child filter because that API is not very clean, and it's not done.

Colleen

00:28:37 – 00:28:47

I'm doing that for the client. But that's what Laravel is. Right? You guys don't have this Laravel doesn't have the concept of child filter, so it's fine. And then you would say, yeah.

Colleen

00:28:47 – 00:28:54

You have to send in the initial query. I mean, it's not rocket science. Right? If we pull those two things out, you're right. It's not what we'd like it to be.

Colleen

00:28:54 – 00:29:04

I'd like it to be self contained, but right now, it works. If they send in the initial query and if we don't allow child filters, so we would just, you know, hide that part.

Sean

00:29:04 – 00:29:29

So so guys, here's my pitch for you to help you feel better about the product that we have now and yourselves. First of all, you guys are both really awesome developers. Really awesome developers. And I think that is what people recognize. And I think that as an early an early adopter of our query builder is gonna be somebody that knows a lot about us already.

Sean

00:29:29 – 00:29:51

It's gonna be somebody that's, like, has this pain pretty deeply, recognizes that. And this has been true of every customer we've had thus far. Right? Recognizes that we have been thinking about this way more than anybody else and recognizes that we are very good at what we do, and we'll we'll get it all sorted. They're going to be willing to put up with imperfect APIs, lack of documentation, etcetera.

Sean

00:29:51 – 00:30:06

Right? They're gonna be willing to put up with that because they get to work with us. They get to work with you. And it is the combination of your skills and the fact that we understand their problem and the fact that we think about it more than anybody else, which makes the product awesome. It's that.

Sean

00:30:06 – 00:30:28

It's not that it's, like, polished and perfect for now. That's the phase we're at. We will get out of that phase. But first, we gotta pull them in because they're gonna help us so much, like, to shape our priorities and, like, figure out what we need to do. Like and there's no possible way that we'll have something that either of you feels good about to launch.

Sean

00:30:29 – 00:30:46

It it just won't happen. And, like, like, as a product. And I think that's also incredibly normal. Like, it is part of the deal of, like, you feel like this is not good enough, and you have to figure out a way to do it anyway. You know?

Sean

00:30:46 – 00:31:01

Like, that is, like, such a common experience. I had that experience with both of my products. It got better the second one. It got better, you know, as I started doing more things. And as I've launched products for clients, etcetera, like, I've had similar similar feels, but it gets better and better and better over time.

Sean

00:31:01 – 00:31:08

It's just one of those things. Like, public speaking, like, yep. It's terrifying, but you're gonna do it anyway. Right? So, like, that's kind of the thing.

Sean

00:31:08 – 00:31:18

And your the little voices in your head try to convince you. It's so hard that, like, this is a mistake. I'm gonna be embarrassed. This is gonna be I'm gonna be ashamed. Like, people are gonna hate me.

Sean

00:31:18 – 00:31:47

Whatever the voices say to your head, like, this is something that we have to learn to ignore because we have to get to the first phase. And the way that I did it one thing that made me more comfortable was okay. Like, for example, you had this happen with Torchlight too, Aaron, with sketching with CSS. All I did was put up the pay like, the ability to pay, like, one day before I was gonna do my actual launch. And, like, somebody found it somehow and figured out how to buy it from me.

Sean

00:31:48 – 00:31:59

And that's the thing. Like, the book wasn't even done. It was, like, you know, 2 and a half chapters and an outline. And I was, like, this is what I'm gonna be doing. And somebody went through all of that trouble to buy it.

Sean

00:31:59 – 00:32:19

Those are the people that we're gonna be working with, you know? So, like Yeah. That's what they want. They would really like us to have a thing that they could use already and they don't care if it's like and so for those people, we're gonna be kicking ass. And then we can like gradually increase our comfort zone to add add more and more people.

Sean

00:32:20 – 00:32:21

That's my pitch for you guys.

Aaron

00:32:22 – 00:32:23

I received this pitch.

Sean

00:32:24 – 00:32:25

Okay, good.

Aaron

00:32:26 – 00:33:04

I, I think you're right. I think there's a, so I think there's a a balance, and the reason your pitch is important is because I'm on the other side. I think so, like, if there's a spectrum or a pendulum, I'm very far on the side of it needs to be of the highest quality before we release, and you're trying to pull me towards the middle. I think if I were on the other side of, like, who cares what the quality is? Let's just shit let let's just ship something.

Aaron

00:33:06 – 00:33:33

That would be, your pitch wouldn't land. Does that make sense? Because I think you're trying to move me towards the middle because I'm on the right side. If I were on the left side and I see a lot, and this is part of my, like, this is part of what has formed my opinions is I see a lot of people on the left side of this fictional spectrum saying it doesn't matter what the quality is. You just have to get something out the door.

Aaron

00:33:33 – 00:33:40

And I feel like that is a bad mental framework. Does that make sense? It's

Sean

00:33:40 – 00:33:43

psychological. It makes perfect sense what you're saying. So I'm trying yeah.

Aaron

00:33:43 – 00:33:44

I'm trying to say like

Sean

00:33:44 – 00:33:45

You are on the good side.

Aaron

00:33:45 – 00:34:02

Depending depending on where you are on the spectrum, that is that affects or that informs the advice that you need to hear. Because I'm so far on one side, the advice I need to hear is you gotta move towards the metal.

Sean

00:34:02 – 00:34:16

Sure. So okay. I you're totally right. I think you're on the my value judgment is that you're on the good side, just too far on the good side. The the other side, though, the pitch for them is different than the pitch for you.

Sean

00:34:16 – 00:34:37

It's a lot different. It's not that, like, you know, the pitch for them is the the reason that people do that shit is because they're trying to get away with something. They're trying to, like, trick the world into making money. They're trying to, like, figure out what's the cool little trick that I need to hack that I need to do in order to make more money. It's not about the customer then.

Sean

00:34:37 – 00:35:02

They've completely left the customer out of the equation. So when you go really far to that end, it's just straight up about them and being selfish. And then it again, on the other side of the spectrum, it's the same thing. I mean, it's it's it's in that it's about you. And when you go too far because then becomes it's about you because you're worried about what the impact is gonna be on me if I release something that's that's bad.

Sean

00:35:02 – 00:35:43

So the middle is like the the egoless zone of, like, I am thinking about my customers as deeply as I can, and the and that informs my decisions. And so part of that is when you get there, then you realize, like, it is actually a disservice to my customers for me to be focusing so much on making it the absolutely perfect thing because the thing that I have now, they could use now and, like, I'm depriving them of that by being worried about it. Like, that's the that's the thing. So either way, and I think that, you know, like I said, my value judgment of this side is way better. Like, this is this is less bad of a problem.

Sean

00:35:43 – 00:35:50

This is more relatable to me. But it is still a problem, and it is still, like, forgetting about our customers.

Colleen

00:35:51 – 00:36:06

Okay. So practically speaking, with the customers that you're you're onboarding White Glove, are you charging them? Or are you saying, hey. If this works for you, I'm going to charge you? Because with Jesse, we were kind of like, this is what we have.

Colleen

00:36:06 – 00:36:18

If it works for you, you should pay us. If not, nah. I mean, so how do you because I've had you know, I do I did have someone reach out and be like, hey. I wanna use it. So what is the process there, do you think?

Sean

00:36:19 – 00:36:20

But we're charging. Right?

Aaron

00:36:20 – 00:36:21

Yeah. That's the question.

Colleen

00:36:21 – 00:36:22

What are you doing? I saw

Aaron

00:36:23 – 00:36:41

you what I've communicated to Neil. So we have we have two cases here. 1 is Keith, who paid the very first licensing fee in 2018. He paid us $500. So he's out, because that just, that's just way outside the bounds of normalcy.

Aaron

00:36:42 – 00:37:08

So with Neil, what I told him was you'll be paying for the license. So you'll be paying $1,000 a year. I am going to spend time for free helping you get it integrated, answer your questions, basically do the the white glove part of the onboarding. So for Neil, he will pay $1,000, and that's it. He's also gonna get a lot of my time for free.

Colleen

00:37:08 – 00:37:09

For free.

Aaron

00:37:09 – 00:37:24

Which will change in the future. People will pay $1,000, and then if they wanna pay for my time, they're welcome to. But I basically, the pitch for, be a beta user is I'm gonna help you a whole bunch

Sean

00:37:24 – 00:37:25

Exactly. For free.

Colleen

00:37:25 – 00:37:27

Yes. Right. You get the free consulting on

Sean

00:37:27 – 00:37:28

top of the You get us.

Colleen

00:37:28 – 00:37:30

You get us. So question

Aaron

00:37:30 – 00:37:45

about I might let I might let him, you know, take 10 days while we're doing the integration stuff and just give him a free license or whatever, and then say, okay, we we got it working. Now you can pay. But I've set the expectation upfront that he's gonna pay for the license.

Sean

00:37:45 – 00:37:59

There should always be special customers that we have this relationship with always because then we have this, like, kinda safe zone of, like, hey. Here's this new thing. I mean, and and it will be generated because it will be like, hey. We're launching this new feature. It's API breaking change.

Sean

00:37:59 – 00:38:12

Who wants to be the first to try it? Those people will be then that special class customer who is like, yep. You get all of our attention because you are helping us to figure this out. Like and that's a win win for everybody. Like, they like it.

Sean

00:38:12 – 00:38:14

We get their we get to work with them.

Colleen

00:38:15 – 00:38:25

So, Aaron, did you demo it for Neil? Like, what is your he has reached out you based on your tweets. And he's like, I want this thing. Did you do a sales process with him?

Aaron

00:38:25 – 00:38:53

No. Neil has been publicly and privately tweeting at me for a while just being, like, number one support guy. Just always being excited. I've especially with the screenshot stuff, that I've tweeted, over the past, you know, couple years, he's always responded and said this looks great. I wanna be the first person to buy.

Aaron

00:38:53 – 00:39:05

Like, let me know what I can do. I'm happy to, you know, pay for it. So he's just always told me, I'm ready to buy whenever it's ready. And so when it was time, I DM ed him, said, surprise. It's ready.

Aaron

00:39:05 – 00:39:20

And then we hopped on a call and he showed me his app, and how he thought he was gonna use it. And then I showed him some of the code and, like, how the, you know, the the interfaces worked. And then he was like, great. Let's set up a time to do it. Okay.

Aaron

00:39:20 – 00:39:28

So that that's how it worked. He was super warm in that he kept approaching me and saying, please take my money.

Colleen

00:39:29 – 00:39:49

Okay. So other practical considerations we have on the rail side is we don't have a way to distribute it as a private package. Not that that's a deal breaker right now. I'm just trying to think if, you know, as we go forward, I mean, maybe I get, you know, like I said, touch base with Jesse. Maybe get 1 or 2 more people.

Colleen

00:39:51 – 00:39:59

Just some things to I guess you're right, Sean, though. Like, these are details that we can work out when we're white loving people. This stuff doesn't really matter.

Aaron

00:40:01 – 00:40:05

So two thoughts. 1 is we need to talk to Filo Filo.

Colleen

00:40:05 – 00:40:05

Right.

Aaron

00:40:08 – 00:40:24

Oh. I I don't know how you pronounce it. But, the guy in our Slack, the the guy that runs unlock sh Mhmm. Because he was investigating, Ruby at some point. And I think Pete actually pointed him in a helpful direction.

Aaron

00:40:25 – 00:40:54

So maybe I and I think he's planning a product hunt launch tomorrow, actually, so now may not be the best time. But I can reach out to him and just tell him, hey, it's going to be like a month, and then we're going to want to sell Ruby. Are you gonna be there? Because, man, it would be the best thing in the world to keep all of that under one roof. Because he handles payment and he handles distribution and licensing and everything.

Aaron

00:40:54 – 00:41:09

And that would be amazing. So that's number 1. I'll talk to him, and see if he thinks he'll be ready. Okay. Number 2 on practical considerations for Ruby is and I'm about to blow your mind.

Aaron

00:41:10 – 00:41:21

The Laravel version now can accept a query from the outside world. So, the initial query can come from outside in the Laravel version.

Colleen

00:41:21 – 00:41:24

Yeah. Dude, I would throw something at you if I had

Aaron

00:41:24 – 00:41:49

something at the bottom. Glad I'm safely on the other side of a screen. 3,000 miles away. So listen, before you freak out, listen, there is, in Nova, there is no way. So the query is generated by Nova and then given to you as the developer to do whatever you want with it, but you can't replace it.

Aaron

00:41:49 – 00:42:08

And so no matter what I did, I couldn't, I couldn't use our initial query. Yeah. I had to receive the initial query that Nova had generated before I had control of the world, if that makes sense. So it runs through Nova's middlewares, and then they hand it to you and you hand it back.

Colleen

00:42:09 – 00:42:09

Okay.

Aaron

00:42:10 – 00:42:19

So practical consideration for Ruby, I use the past in initial query if it exists. And if not, I use the one defined in the filter. So That's

Colleen

00:42:19 – 00:42:21

what we have right now in Ruby.

Aaron

00:42:21 – 00:42:50

So there you go. You can just basically not tell other people that they need to pass the initial query in, and nobody will be anything the wiser. So, oh, and 3rd or to be, because I said 2 things to be is, I feel great about shipping Laravel right now. Like I have reached, I have reached the point where I think this is ready to go. So it may appear, as some total asymptotic, it may appear like an asymptote.

Aaron

00:42:51 – 00:43:12

Like I'll never reach the point where I'm ready, but I'm, I'm totally ready. And I think this is, I think we're in a very good spot and I feel comfortable launching it. So the argument could be made, you know, we should have been there 6 months ago, but regardless, I feel great about it. I think we're totally ready and I don't think we should, and we're not delay any further. So it does happen.

Aaron

00:43:12 – 00:43:14

I I do get there eventually.

Colleen

00:43:15 – 00:43:27

Mhmm. That's awesome. Yeah. I think for Ruby, the steps are I think, Sean, you and I should try and put Vue together with Rails and get that working. Like over the next I mean, I don't know.

Colleen

00:43:27 – 00:43:51

I can't do it this week. But like over the next month or so, we'll just schedule some time. I think we should do view with Rails and then pull out all of the front end stuff that we're using with client, and get that working, package it as a gem. And then as as I said, for now, we'll just send in have them send an initial query. If they have that tenancy issue, which, honestly, some people don't.

Colleen

00:43:51 – 00:43:59

I mean, again, it's like, this isn't even an issue for everyone. And, and yeah. Let's do it. Okay.

Sean

00:44:01 – 00:44:08

Yeah. So if anybody reaches out to you or if they have reached out to you, now would be the time to start, like, lining them up for integration.

Colleen

00:44:09 – 00:44:11

On it. And

Sean

00:44:13 – 00:44:14

put it on the calendar.

Colleen

00:44:15 – 00:44:16

Put it on the calendar.

Sean

00:44:17 – 00:44:21

Yeah. When we're gonna integrate with them. Because if it's Oh, yeah. If it's possible. Right?

Sean

00:44:21 – 00:44:27

Collect the money, put it on the calendar. It gives us a constraint of time, which is very helpful.

Colleen

00:44:28 – 00:44:31

Yep. Yep. I think so. That sounds good.

Sean

00:44:31 – 00:44:31

Okay.

Aaron

00:44:35 – 00:44:39

Sean's excited for customers. I'm excited for that sweet, sweet license money, baby.

Sean

00:44:40 – 00:44:41

Well, I'm

Aaron

00:44:41 – 00:44:42

excited for that too. We're all gonna retire.

Colleen

00:44:44 – 00:45:02

Dude, Aaron, let me tell you the story. So I was Sean was here, what, 3 weeks ago, and we're having lunch or breakfast, and I'm telling him about these some of these big life decisions my family has to make. Mhmm. And he looks me dead in the eye, and he's like, man, your life would be easier if you were rich. Yes, sir.

Colleen

00:45:02 – 00:45:06

Yes, it would. It was awesome. It was a moment.

Aaron

00:45:06 – 00:45:09

I don't even need to know what the decisions were. Of course, it would be.

Colleen

00:45:10 – 00:45:11

Yes, it would.

Aaron

00:45:13 – 00:45:16

Yeah. That's amazing. I

Sean

00:45:16 – 00:45:18

don't even remember saying that.

Colleen

00:45:18 – 00:45:19

That's Oh, I remember that. That

Aaron

00:45:19 – 00:45:25

sounds right. It sounds right, though. Who has something else?

Sean

00:45:25 – 00:45:41

Yeah. Marketing. So I started working on first step with the marketing is we gotta have ourselves set up so that way we can get the so we can write our own landing pages, do our update our okay. Wait. Step back.

Sean

00:45:41 – 00:46:31

1st step I did was I got with Buckbee and talked about SEO. So the what what I did with him is we looked at all my sites, all my content on my planning for aliens business side and figured out what of those things was valuable to us and came up with a plan for, like, structure for how to structure the new hammerstone.dev site and how to you know, strategy for moving that over. So phase 1 of the new marketing site will be the updating it to the new system, which I'll come back to and like how we're how we're doing that. And then getting all of the SEOs, like, juice from my sites pointing over to there. And then, also, I'm gonna see what I could do with my mailing list.

Sean

00:46:31 – 00:46:44

I mean, it's cold. It's dead. But there's 20,000 people on there, and surely some of them remember me and might care about this. So I'll see what I could do there. The so that's that's that.

Sean

00:46:44 – 00:47:13

So phase 1 will be the new site. It will not be, like, designed yet. It will be just something, but it will have all of the SEO stuff over and it will be using the new system. And all of this will be done before Aaron goes on Laracon does his presentation as the timeline. So then phase 2 will be an actual, like, landing page for specifically, I'm gonna focus on refine.

Sean

00:47:13 – 00:47:42

So we'll have a good product landing page for that. And then I'll probably go with Bugbee again and sort out, like, kind of, you know, best guess at, like, a flow that we should have. So, you know, what what should be on the home page, how it should go from there. I'll probably work with everybody in b biz like I usually do on that stuff, and we'll have that. So that'll be phase 2 will be like, it'll be nicely illustrated and designed and look good and be like a and be like actually effective marketing copy and try to do some jobs

Aaron

00:47:42 – 00:47:43

for us. Lyricon, right?

Sean

00:47:43 – 00:47:45

No, no, no. Before Lyricon.

Aaron

00:47:45 – 00:47:50

Phase 2 is also pre Lyricon. Okay. Which is February 9th for those who don't remember.

Sean

00:47:50 – 00:48:01

So so that's phase 2. I just wanted to let you know that it's going to be broken into 2 phases because there's going to be a phase where it's going to look not great, but it's going to be a website that has all the SEO stuff happening.

Aaron

00:48:01 – 00:48:10

Great. And will that, I guess, will that include docs for our current packages? Because I think that'll be important. Okay.

Sean

00:48:10 – 00:48:31

So then alright. So the new system. So my pipe dream for marketing platform has always been, like, I just want to write a markdown document and put it in the folder that I would like it to be and then have it show up, like, on the website nicely formatted and boom, I'm done. And, like, I can just use GitHub to be, like, git commit push, and then it's there. So we're gonna have that.

Sean

00:48:31 – 00:48:37

I got that. I'm, like, halfway to getting that set up. So I'm doing it on my site first.

Colleen

00:48:40 – 00:48:41

What are you using for that?

Sean

00:48:42 – 00:48:44

I'm using that thing we talked about before called remix.

Colleen

00:48:45 – 00:48:45

Okay.

Sean

00:48:46 – 00:49:11

So basically, the way it's gonna work is you write is it work exactly like I said, you write a markdown file, you put it in the folder where you want it to show up and then you get commit and you push it and it shows it shows up. And the way that all works is, it will be completely deployed on the edge. There will be no servers. The remix app runs completely on its own on the edge. The data will be stored in Cloudflare's key value store.

Sean

00:49:11 – 00:49:41

So when you push to GitHub, if it's only a content change, it will just run a GitHub action, generate the content from the markdown, and then, put it in the key value stores. And then the Remix app will just load it from the the updated values. It's basically actually how the remix document documentation site works right now. Then far as our docs go, there's the extra step is it's gotta go through torchlight, but that's like a separate, like, server side call. So that should work fine.

Aaron

00:49:42 – 00:49:57

Question. For the markdown, put it in a folder and it shows up. What kind of content is that for? Because, like, I'm thinking landing pages that won't work because they need to be, you know, super nice and designed and whatever. Is that specifically for the docs?

Aaron

00:49:57 – 00:50:00

Or are we having some other kind of content in the middle there?

Sean

00:50:00 – 00:50:04

Yeah. That's for docs. That's for how to's. That's for think pieces. That's for cheat sheets.

Sean

00:50:04 – 00:50:16

That's for for that kind of stuff. But we also have the ability to override that completely and do something in the Remix app itself. So it's JSX, and then you can do whatever the hell you want. Like, it just yeah.

Colleen

00:50:16 – 00:50:21

So were you taking the docs Aaron has already written? Are those in markdown, Aaron?

Aaron

00:50:21 – 00:50:22

They are.

Colleen

00:50:22 – 00:50:22

Okay.

Sean

00:50:22 – 00:50:24

It should just work.

Colleen

00:50:24 – 00:50:27

Okay. So I can write all the Ruby docs in markdown is what you're telling me.

Aaron

00:50:28 – 00:50:28

Yep. Yeah.

Sean

00:50:28 – 00:50:34

Use Torchlight or whatever Torchlight needs in order to know code syntax highlighting is.

Colleen

00:50:34 – 00:50:36

Perfect. That's good news.

Aaron

00:50:36 – 00:50:38

Cool. That's exciting.

Colleen

00:50:38 – 00:50:41

Yeah. That's awesome. And that's gonna be done in a month?

Aaron

00:50:42 – 00:50:46

Yeah. I know. That's what I'm thinking. Rock on. Works for me.

Sean

00:50:46 – 00:51:16

I'm mostly I'm mostly there. I mean, I got the plan and I'm mostly there with the, with the remix, implementation for my site. I'm doing it my site first. So that way, if it's crappy, I could do kind of like a crappy version on learn on my site and then do it for us. So like ours is going to be different cause like on mine, I'm just going to have like one folder for content slash blog and only have posts in there versus I'm going to do like a little bit extra fancy stuff and have like slash there's going to, we're going to have like slash blog slash docs slash we'll probably won't have slash blog.

Sean

00:51:16 – 00:51:49

We'll have like how tos or whatever. This is the SEO conversation with Bugbee. And then, and then we'll have, the ability to have like nested folders and stuff like that, which is not too much more work, but that's just like some tweaks on the, the code that I'm writing for the slightly simpler version on my blog. But yeah, as per usual, the thought the hard thing is not the code. It's figuring out all the integration stuff with like CloudFlare and the key value store and like get it set up so it works in development and then get it set up so it works as Git actions and all that crap.

Colleen

00:51:49 – 00:51:49

Yep.

Aaron

00:51:52 – 00:51:57

Yeah. Pay close attention while you're setting it up locally because I've never used remix or anything like it.

Sean

00:51:57 – 00:52:01

I'm gonna write some content about it. I've been taking notes. So that'll be

Colleen

00:52:02 – 00:52:02

Okay.

Sean

00:52:02 – 00:52:17

Because I think that's also good content to write too because they've got a lot of momentum right now. They do. So, and if I can do something, do something cool with remix then, or even just a basic how to at this point for remix would not be useless.

Aaron

00:52:17 – 00:52:26

And then whenever I, whenever you're done with that and I figure out the Torchlight integration, that's even more content for, for the site as well.

Sean

00:52:26 – 00:52:30

The Torchlight integration for this? Yeah. I don't think you will have to

Aaron

00:52:31 – 00:52:34

I I got to hook it in somewhere to the markdown processor somewhere.

Sean

00:52:35 – 00:52:41

So it's just an API call, right? Like I just make a get and then send it our doc and then you give me back the annotated doc.

Aaron

00:52:41 – 00:52:48

Yeah. But I'll need to hook into your markdown pipeline at some point to grab the blocks and everything.

Sean

00:52:49 – 00:52:52

Sounds like it'll just be like one code, one line of code, because it'll be

Aaron

00:52:52 – 00:52:53

That would be amazing.

Sean

00:52:53 – 00:53:10

Because, cause all I'm doing is like just going through oh, because the way to work is I'm gonna get the GitHub tarball from Git and then, you know, expand it, go through the files. And then for each file, run it through the markdown processor. And then we just add one more step. If it's docs, send it through Torchlight.

Aaron

00:53:11 – 00:53:14

Cool. Yep. And that'll be great.

Sean

00:53:21 – 00:53:23

Alright. That's it. That's the marketing update.

Aaron

00:53:23 – 00:53:38

Nice. I like it. Mhmm. Update on my side, besides the stuff I already talked about. The so Laracon is February 9th as we've as we've said.

Aaron

00:53:39 – 00:54:19

I recently so what I'm presenting on there is Sidecar, which is the, Lambda wrapper, basically. So it lets you execute Lambdas from your Laravel app. I did some work on it this past weekend, right before I came back to real work. There were a couple of open issues that I finally had a chance to look at, fix some of them In the process, I saw an issue from somebody that was like, hey, it would it be possible to, run PHP in Lambda? And it's possible, but there's no native Lambda runtime for PHP.

Aaron

00:54:20 – 00:54:41

So you can't just like pass in the string that says, you know, PHP 8.1, because Lambda says, I don't know what that is. Whereas with node, you can say, I want node 14 and they say, got it. So Sunday night I stayed up way too late. I was, you know, it was basically like holiday revenge. Like I don't want to go back to work.

Aaron

00:54:41 – 00:55:25

It's still the holidays. So I stayed up till, you know, 12:30 or 1 on Sunday night and figured out how to run PHP on Lambda, while also not having to maintain a PHP runtime. So and then this is the thing that I tweeted about earlier. I think it was yesterday or the day before, and I posted it in our, you know, tiny wins channel because it's gotten, like, 230 likes and, you know, couple dozen retweets. And the thing that I figured out and and this is gonna be great because I'm gonna present this at Laracon, and it's gonna blow people's minds.

Aaron

00:55:25 – 00:55:57

But the thing that I figured out is I can use Laravel Vapor, which is Laravel serverless platform. I can use Laravel Vapor's, PHP run times because they're publicly available layers. So I can use their PHP run times, and then I can ship, I can ship a closure to Lambda, run it on Lambda, and get the result back all from my local machine. So it's wacky. It's totally out there, and it's incredibly cool.

Aaron

00:55:57 – 00:56:54

And so I don't really know what people will use it for. I mean, this person who asked for it on on the repo obviously has some ideas. But this morning, I actually got it to where I can ship my whole Laravel app to Lambda and run a command on Lambda within my app scope from my local machine. So we're getting to the point where I'm gonna create a black hole at some point because it's just like total inception. But now, what what that means is you can, from your, let's say, from your production web server, from your local machine, or whatever, you can spin up a 100 or a 1000 copies of your app on Lambda and run a 100 or a 1000 commands all in parallel, and then get the results back to your, you know, your main web server or whatever.

Aaron

00:56:54 – 00:57:35

So if you had something that you needed to do in a in with high parallelization, you could just send it to Lambda, let Lambda run it all and get all the results back. And so I'm kinda pushing the limits of what is, like, theoretically possible here, but it's amazing. And I think this is gonna go really well at Lyricon if the Twitter response is any any indication. So feeling kinda pumped because I didn't really know what I was gonna totally present at Lyricon. I have, you know, stuff to talk about with sidecar, but this is gonna be the this is going to be the one more thing, in the Steve jobs style, that's going to blow people's minds.

Aaron

00:57:35 – 00:57:39

So we'll see, but pretty excited about that.

Sean

00:57:43 – 00:57:46

That's going to be fun, man. Can't wait to see the response to that.

Aaron

00:57:46 – 00:58:01

I know. I don't know what people, I really don't know what people are going to do with it. But people are saying they've got ideas for it. It's kind of weird to be like, I'm building this thing. I don't really know why it's gonna be useful, but I think it will be.

Aaron

00:58:02 – 00:58:05

So we'll see.

Sean

00:58:06 – 00:58:07

Alright dudes, could we wrap?

Aaron

00:58:07 – 00:58:08

Yeah. Let's wrap. We gotta

Sean

00:58:08 – 00:58:12

have our conversation after this and then I got work to do. K.

Colleen

00:58:12 – 00:58:12

Back in the Saddle

Transcript

Not enough emails?