In this episode, Tom Zaragoza and Charlie Reese talk about when it makes sense to use lower level, customized code (i.e. yes code), no code tools (i.e. no code), or low code (i.e. high-level libraries).
In this episode, Tom Zaragoza and Charlie Reese talk about when it makes sense to use lower level, customized code (i.e. yes code), no code tools (i.e. no code), or low code (i.e. high-level libraries).
00:00.00
charliereese
Welcome back to another episode of the deaf printur as always I'm joined by Tom Let's going out Tom same old same old same old.
00:06.64
Tom
Um, hello same old same old same old.
00:36.78
charliereese
Just another day another another saas project and today we're going to talk about a um, an argument arguments probably there. Ah the wrong word, but we're going to talk about. Ah.
00:33.56
Tom
Yeah.
01:10.82
charliereese
What I call yes code versus no code versus low code which is best for what circumstance when you might want to get a very kind of customized and low level code. When you might want to reach for no-code tools and when you might want to reach for what I call a low-code tool and an example of that would be if it was for backend software a framework like Larell or rails where it does a lot of the heavy lifting for you assuming you're okay. With kind of like best practices essentially so why don't we kick off Tom so yes, code no code low code when do you think each makes sense and who do you think. They make sense for.
03:00.10
Tom
I Am if you are building Saas I I think I'm always going to be team yes code and then as you realize you don't need certain things you can start to take them out and then you can go the no or low code. Um, but like that's just me. Um I think you know what I mean like it's It's just because like I'd rather there's two things. There's there's there's 2 sides right? One side is like do I want to bang my head against the wall because code doesn't work.
04:16.12
charliereese
Um.
04:13.30
Tom
Or do I want to ba my head over against the wall because the thing that I thought like no code or whatever was going to be was going to help me but like instead. Ah I'm just building workarounds for it right? So it's like I'd I'd much rather bag my head on the wall for code.
04:50.68
charliereese
Yeah I hear you and you're you're a developer so that makes a lot of sense right? as am I as am I that that relatively closely aligns with how I feel and to expand on that I would say this so for yes code I think it makes sense for 2 kinds of people.
04:48.54
Tom
Because like I know that it was yeah totally. Yeah yeah.
05:30.18
charliereese
Um, companies that are preparing to scale or have already scaled with good reason so they have a relatively um, they've they've collected a large amount of value which is probably in the form of sales so they're generating cash. But in some instances it could be a company like Youtube before it put ads on the platform that was just very valuable and needed to scale so makes sense for those companies that like actually need to scale. Not not companies that might eventually need to scale the companies that actually need to scale the other I think good use of you.
06:26.68
Tom
And.
06:43.30
charliereese
But I'm calling yes code um is new software developers I think it's really important when you're a new software developer to try to do as much as possible yourself and to use as few Frameworks as possible. So you understand how basic things work.
06:37.46
Tom
You get this.
07:20.64
charliereese
I remember when I was learning how web applications worked I spent some time working directly with requests and responses with very minimal frameworks involved. Um, and I also built my own web application framework at None point just so I could kind of understand how like the rope matching worked. And going from like model to view to controller works and all of those things. Um I think it's very very helpful to do the full-on code thing if you're learning. Not only if you're preparing to scale with good reason or scaling. Um no code who is that for. I do pretty strongly feel that there are none good uses for no code the first use that I think everyone would probably accept is non-technical founders that are trying to roll something out um to test it. Um, and the second use and there'd be a lot of overlap with the first use is every so often you find a tool that doesn't require any code that functions very very well and there is no need to heavily customize it or there is a very rich plugin ecosystem already available. A good example of that would be wordpress. Um, even though you can add code it. It is largely a no-code tool um or a blogging platform like ghost. Um I'm trying to think of another good example of a very good, no code tool. Where you almost never need to inject your own code to customize it unless you're doing something crazy Tom can you think of one like maybe a shopify store that's probably a good example right.
10:25.14
Tom
Yeah, yeah, that's that's probably a good one I mean I'm kind of hard pressed to think to think of one just because of you know, yes code. But I think shopify is like a really big idea.
10:50.24
charliereese
Yeah, yeah, so so so there definitely are some examples where the no-code tool has gotten It's both um it it is niche enough that they've been able to kind of define exactly what the tool is for. Um, and it is developed enough that it can do kind of 99% of mainstream use cases and I would say shopify. That's the case wordpress that's the case wordpress is for blogs. It can do most blogging things you would want shopify. It's for ecommerce stores. It can do most ecommerce things that you would want. Um. Where I think low code comes in is anything outside of yes code and no code and so yes, code being you're trying to learn more about software development or you are building something that actively needs to scale. Um. And then no code being you either don't know how to code and you're trying to found something and you're trying to spin up something relatively quickly or test something relatively quickly. Um, or there is a product that fits your use case very very well that you will not need to customize and so low code and I mean our podcast is about. Growing a kind of what will probably be a side or hobby saas project from None None $10000 um mrr um I think 99% of the time low code makes sense for that and what I mean by low code is you have full access to code you are writing code but you are making heavy use of what. Is probably open source software and in some cases paid software um to accomplish what you're trying to accomplish but you still can crack everything open and completely customize anything you need. Um, that's what I think makes sense Tom Tom what do you think about.
14:11.60
Tom
Yeah.
14:23.22
charliereese
Those caveats around yes code, No code low code. Do you think those are fair.
14:20.48
Tom
Yeah, no I think they're fair. Um I Love your point about no code being. It's it's something that has really narrowed in or honed in on like a specific use case right? like shopify for stores right? Like you know it's a store like the the.
14:57.10
charliereese
A.
14:59.24
Tom
Flows and the processes there are is pretty pretty standard if you will um so I subscribe to that and it doesn't make sense really to build out a whole store or to to build it a whole blog if you have solutions like that. Um, but.
15:31.64
charliereese
Yeah, and and it really is a time thing right? So so when it makes sense absolutely because you save the time but that's why I think no code or sorry low code often makes sense because again you're saving the time and you're not necessarily trading anything away for that because it's ah like.
15:36.60
Tom
Yeah, just.
15:57.18
Tom
Brother.
16:10.68
charliereese
Similar to no code low code to me is it's a more solved problem area but not perfectly solved and so they're able to mostly solve the problem for you but then let you inject whatever code, you need to solve it. However, you want to solve it and so a good example of that.
16:24.14
Tom
Um, right.
16:47.80
charliereese
Would be a frontend or a backend software framework where you can still completely customize it and you can get it to do whatever you want, but it does do a lot of the heavy lifting for you? um. And it's obviously a spectrum like it's not yes code no code or or low code like I would say a framework like Express in Javascript is somewhere between low code and yescode because it doesn't do as much as the heavy lifting as rails does and you you need to take care of a lot of that yourself. But it still is layering in a lot of functionality that you now don't have to worry about but I would put it more on the yes code end of the spectrum because very few people professionally are doing absolutely everything themselves. Um, which would be kind of the furthest you could go on the yes code spectrum is that fair.
18:11.60
Tom
Right? one? Yeah, that's totally fair. Um I like you know the whole spectrum from yes to no to low code one. So when you so when you say low-code. You know you're talking about like a framework right? You're talking about like rails and um, you know.
18:56.60
charliereese
Yeah, or even something like ah and this would be very low code. It's literally like you're injecting a couple lines of code but something like Google analytics right? where you're injecting a couple lines of javascript to track.
18:50.52
Tom
Like maybe next one.
19:32.50
charliereese
Page visits or whatever you're trying to track. Um I feel like that that's getting very close to low code I would actually say but it's somewhere in between no code and low code.
19:41.34
Tom
Got it one use case that I've I think sort of falls in between no and low Novo is um, headless or like so you know jamstack right? So you have on the front end for example, next.
20:05.00
charliereese
Um.
20:23.66
charliereese
Um, yeah.
20:20.68
Tom
Um, which you know it's content right? But instead of you building out an entire Cms you can tap into. Um you know like that's kind of where like code comes in right? like you could tap into the api of say wordpress or ghost. But then you can have full control.
21:00.30
charliereese
Yeah, but I mean you're tap it into an Api right? like I feel like at best that's low code. Not no code and you're still writing potentially html with some of like the jamstack stuff like it's not.
20:59.60
Tom
On the front end.
21:07.76
Tom
Um, but then sure.
21:27.82
charliereese
1 line you're injecting somewhere like you're still doing it. You're just using some of the heavy lifting that can be accomplished by frameworks to me that is a very good example of low code.
21:40.52
Tom
Sure. But then what about the side where it's no code where you know you're you're using a Cfs So instead of you actually building out.
22:00.88
charliereese
But so so using a cms isn't necessarily no code I think using a cms through a Ui or no code is required is no code I think the distinction is no code, not not whether it's a system that's been created by someone else if if you need code to interact with it.
22:09.82
Tom
Um, yes, and that's right Gail.
22:36.90
charliereese
I Think it would still count as low code.
22:34.52
Tom
So um, the Cms right? I'm just I'm just trying to figure out where on the spectrum this kind of would lie because on the front end. Yes, you'll have next and it interacts with the backend sort of Api to to get the content but that's a right click.
23:07.56
charliereese
You know and just for their record anyone listening Cms content management system. Most people probably know that next is a Javascript framework for um. Using Javascript on the backend to basically render frontend components and send them to your browser if you're like loading a website was just just wanted to provide context for anyone who is not technical.
23:54.58
Tom
Um, so yeah, yeah, yeah, cool cool. So my I guess where I'm saying like it kind of falls in the spectrum as is yes lowcode you have another frontend. It's interacting with the backends getting the information but then the no code side of it is. You're using a Cms.
24:29.10
charliereese
And.
24:34.46
Tom
You have like the Ui you log into ghost you log into wordpress. Whatever you're you'relying on using but instead of you actually building out a full on. Um you know cms but you have full control over how you how it looks on the front end.
24:52.18
charliereese
Yeah.
25:08.38
Tom
That's where it kind of like touches on or has its foot in no code side of things where it's like you don't need to build a Cms.
25:19.64
charliereese
So so I agree except as long as you're hitting their api. That's not a no code thing right? It's just low code. It's just you' borrowing a framework that's already been created I think it's no code if there is no code right? You're completely using a Ui to spin it up.
25:46.86
Tom
Um, okay I see it.
25:57.60
charliereese
You're not hitting their Api right? like I think it's all in the name um to me what you're describing is a perfect example of lowcode like you're completely relying on an Api that someone else developed but you're still hitting that api and deciding how to show stuff that comes back.
26:03.50
Tom
Um, okay I said I said.
26:19.56
Tom
Um, fair enough I'll go.
26:32.52
charliereese
Um, but you know what? That's another good example of a low-code tool. It's not only necessarily a backend framework like nextjs but also a head less cms whereas a non-headless cms just using wordpress through your browser would be no code. Using a headless cms and interacting with wordpress through an Api or ghost through an api would be a good example of low code I think that's fair. Um, yeah, so yes, code no code low code I think it's all about saving time when it makes sense to save time. Sometimes yes code you're learning or you're optimizing doesn't make sense to save time low code. Um, you're trying to save time where you can but you still want to have access to customize anything you need to customize you just don't want to reinvent the wheel and so I think for the overwhelming majority of people. Low code is the thing that makes sense and then no code I think there's a couple use cases. But basically you, you will not need to customize anything or you don't have the ability to customize anything. Um. Or saving time isn't exceptionally important and you just want to fire something out as fast as possible. Although I do think low code fills that void fairly well because it's a spectrum so you can kind of just go far on that spectrum. Um I want to talk a bit a little bit so so I'm working on a project in my spare time. That is going to be software related to mountain biking. Essentially so I've been working on it for I think I've put like around twenty days into it right now and I say twenty days and there have been I think 20 times of like committed code. Or something like that. Ah, and you style have worked on it. It's been like an hour or something sometimes less sometimes little more so so twenty days being like I've worked on it 20 times at some point so I guess not twenty days but I mean however, you want to cap that um and. Tom I was chatting with you a little bit before we started recording this but I realized that on the frontend and for design on the backend and the infrastructure kind of cloud hosting solution I'm using all of them I would say are low code. So I wanted to walk through all of them because I thought it'd be kind of fun and talk about why I picked these low code solutions and what I used to do before that Tom what do you think? do you think that'd be like a helpful way to kind of illustrate. Yeah, okay, so so so frontend and design.
31:42.16
Tom
Theu.
31:59.80
charliereese
Um, I used to spend a lot of time which again as Tom and I discussed I think actually makes sense when I was trying to learn frontend design pardon me when I was trying to learn frontend and design. Um I kind of got in there I was doing the yes code thing. Um I was using tools like sketch a lot to create my own designs. Um and just trying to do all the heavy lifting myself because I was trying to learn how to create good designs and how to create kind of good frontend functionality components. Whatever. Um, and I did learn a reasonable amount while I was doing that but I am not a professional designer and I would also say I'm not a professional frontend developer. Although I do do a reasonable amount of frontend development. Um, so what? I'm using for my new project is tell when Css which is um. Feel like they'd honed identify this way but I would say it's quite similar to bootstrap. Um, although it ah is maybe a little more customizable kind of arguable. But yeah, it's somewhat similar to bootstrap. But I think a little later wait. Um, and they also have a paid so the css is free the Css library but they have a paid ui components package that I think is around none for the entire package and it is a library of professionally designed components um like signed in pages. Landing pages, pricing pages, whatever and and components that you can just take off the shelf when you're building a saas application and customize with your own colors functionality you hook in Javascript do whatever you want to do? um. And that has been fantastic because instead of spinning my wheels coming up with what to be honest are less than impressive designs that have taken me a lot of time. Um, the application has a great design. It looks very professionally done. But it still has that kind of custom branding that I've wanted to inject in with my own colors and logos and whatever and so I've been a huge trend of that and I've saved so much time on the front end and and with design by kind of finding a low code solution where this html and css and kind of this designing has been done for me. And I guess this is a paid example instead of an open source example. Although tell when css is open source is just the ui components are paid and then I can kind of inject my own code just a little bit of code to do what I want so I've been a huge fan of that on the frontend and for design that's been fantastic tail when Css and ui um.
37:12.66
charliereese
Highly recommend. Although the ui is paid for most the components. Um on the backend I've tried a lot of backend frameworks mostly professionally, um and somewhat just for personal curiosity learning what have you I keep coming back to else. I do think that Tom you mentioned nextjs I do find that framework relatively compelling and it's nice to keep everything in Javascript. It's nice to be able to render react components on the backend. Um I find it relatively intuitive the framework but it doesn't come with all of the batteries that rails does. It's just not as batteries included as rails um and the documentation for rails is just so good. Um, it's so stable. There's so many examples of cut companies growing very quickly using it. Um I think part of that is ah. Because you can accomplish so much in the framework so quickly and it's so easy to understand what code has been written and what it does partially because I think of the way the language ruby was designed but yeah on the backend they just keep coming back to rails I love it. I I find I'm so efficient in it. Ah, more efficient than I am in any other language even though I arguably write in other languages far more um and other frameworks. But yeah, using rails been super quick I've been super happy with it. I love the none party like plugin and library ecosystem and ruby and rails. Fantastic if. You haven't checked it today. You're a software developer. Definitely give it a go learn how the magic works get comfortable with it. It's really easy to work really fast. So I've been doing that low code on the backend side of things infrastructure. This is another one where. I mean like all facets of software engineering I started with the yescode thing to kind of understand how things learn or worked um I created a relatively unpopular ruby gem for basically generating terraform files and. Auto deploying your project to Aws. Um, so I went the full-on yes code thing for infrastructure and kind of cloud infrastructure deployment. What have you for a while. Um and I am back to using a very no code or sorry a very low code solution. Which is heroku and it's basically a managed platform as a service um with the caveat that I'm actually using aws for my database and for s 3 and the reason I'm doing that is I didn't know this for a long time I feel like a lot of people don't know this.
42:40.40
charliereese
Heroku is actually hosted on Aws and so if you put aws resources in the same region as your Heroku resources. You basically don't have networking overhead and it's way cheaper and so instead of paying a huge amount for a piece of garbage database on Heroku. I'm paying like 5 or 10 times less for the same thing on Aws um, and it's working great. So I'm doing that for now. Ah still kind of taking advantage of the more expensive but also far easier to deploy. Um. Web application servers and worker servers on Heroku um, but to me it's worth it and you kind of have logs and everything you need over the box and it's so easy to deploy and it's really easy to set up ah cicd stuff with it. Um, can't recommend it enough. Great platform worth the price. Although I'm cheap. So I kind of found a way around some of that. Um, yeah, so it's kind of Tom that's kind of would have been doing on my news ass projects has been low code for everything What do you think about what I'm using on the frontend backend and the infrastructure.
45:06.62
Tom
So if we were to go start with frontends. You know, same thing right? like you don't want to be designing things from scratch right? If we're like in the interest of time and in the interest of eye candy.
45:44.16
charliereese
Yeah, that now with the caveat I would say if you're if you're trying to learn I think that's the only way to do it. Ah but but yeah, once you kind of learn the basics I Totally agree if you're not trying to be a professional designer. Don't don't do it.
45:40.78
Tom
Right? Like you want it to look good and you want it to be as fast as possible like you want to develop as fast, but. Um, Oh yeah, no totally? Yeah yeah, um.
46:23.00
charliereese
Just ah as much as possible, kind of look for existing frameworks and solutions whether they're open source or paid I hundred percent agree with what you're saying. Yeah.
46:33.90
Tom
And when it comes to back ends like people have their preferences based off of languages that they likes to weren't work in so you can't really, you can't really compete with you know, something that something someone's good at using and what they're familiar with. So.
46:54.58
charliereese
I Yeah well.
47:13.26
Tom
You know there's really nothing really to say there unless it's like something that's like totally unusable. Um, but.
47:29.38
charliereese
Yeah, that that one I have some plot takes on I agree that say you want to build a side software project and in a year you'd like it to have like a material amount of revenue. Then yes I fully agree with what you just said use what you know. Find the best possible kind of frameworks and libraries in the language ecosystem you're in whether that's like python java ph p ruby javascript whatever um, then? Yeah I agree my hot take. And so many of you will disagree with this and actually I wrote an article on it that was on the front page of hacker days for three days ah the creator of ruby retweeted it fired a lot of people up a lot of people freaked out but it it was on. Ah. The most successful y combinator startups and what languages they used in their tech stack and to me it was very compelling very compelling that for certain types of software. Certain languages and frameworks are better designed and better able to basically allow you to develop very quickly. Um, and so my hot take is that if you are developing for a web app. There are only a couple. Options that are kind of optimized to allow you to release something good quickly. Um versus other options relatively and I would say the same thing about if you're developing some sort of data science solution. Um, or if you're developing basically some sort of low-level software. Um, there's only really a couple languages and ecosystems that make sense and allow you to be very productive very quickly. So yeah, like while I agree that if if you're giving yourself a year and you want to put something out totally kind of have to use what you know and just pick the best frameworks languages. Whatever. Ah, but you can I do think if you're going to give yourself a little more time and you're willing to learn the tool that makes the most sense. Um I would say that in the web right now. It's going to be either the javascript ecosystem or the ruby ecosystem I really don't feel that there is another. Ecosystem that is as well-represented um with ah relatively if like quickly efficiently built large saas applications I think it kind of has to be Javascript or ruby um, once organizations get to hyper scale. Obviously it's different.
52:48.80
charliereese
Um, they have the resources to kind of build in whatever language they want to scale as best as possible. Whatever but there's just so many examples of companies using rails for instance to get to scale successfully? Um, that is really hard to Ignore. Um and so that's my hot take on that. Ah. I Think the common answer use what you know whatever I think it's It's like it's kind of like a yes, that's true, but um, because I don't think it's necessarily optimal. Um, it just depends what your timeline is.
53:49.18
Tom
Fair I mean at the end of the day right? All we care about is building something that that we can launch and earn money from so the fastest way to that will be using what you know and like if you don't know something that's kind of. Battle tested then by all means learn how to use something that is valid tested right? So your railils and your next or whatever and honestly, that's good on.
54:38.80
charliereese
Yeah, that the caveat I would add to that is the fastest sustainable way. So if you're writing a very small one feature application even spaghetti code is probably sustainable and so it doesn't really matter. But if you're going to build a slightly bigger software application. You kind of need to find a framework that won't devolve into spaghetti code and so in my opinion and this is trigger in in my opinion. Um python isn't as good at web app development as a language like Javascript or ruby.
55:15.36
Tom
Yeah, fair enough and you know most of those.
55:55.50
charliereese
In terms of their best frameworks versus the best frameworks in Javascript and ruby um I mean Phb has Laville now which is very similar to rails. It was modeled after rails I still feel that it has more in direction than rails. Um, it does have some functionality that's compelling that reals doesn't have um but I feel that would produce more spaghetti code. Um, however, obviously that's slightly mostly even just my own opinion. Although I'm sure you could do analysis and show kind of like lines of code per function. Whatever whatever the level of indirection. Ah, but yeah I think what you're saying is fair what you're saying is fair and and you know what we like building and talking about saas applications that are going from None to $10000 um mrr by definition. They're probably a None or 2 feature application right? So by definition. It doesn't really matter if you pick the optimal or one of the most optimal languages and frameworks um, because you're not going to have that much software. Is that fair to say.
58:18.74
Tom
I Would think that applications that can you know that do generate rate that much money in like in that range I don't know man I think just at the end of the day you know rails.
58:42.14
charliereese
Um.
58:57.50
Tom
Laraval Django Next? whatever you pick will do the job and like that's all that matters and like if they need to go past that much if they need to bring on other people to the team if you know those all those other bigger problems like if you get to that point. Then yeah, you can start to think about that or you can start to rearchitecture how you kind of structure the code if you need to um so like at the end of the day. It's just you ship it and then you just get it out there and you just because there's there's so much more to it than just like is the is the model in the correct folder.
59:51.78
charliereese
Yeah, you know what? I yeah.
01:00:19.20
charliereese
I I hear you and yeah and that's not necessarily what I'm saying but like to repeat what you just said as an analogy and I think is easier to understand. Let's say you're building a house right? Some people know how to build with concrete. Some people know how to build with wood. Um.
01:00:15.64
Tom
Or whatever, right? like you shouldn't really be worried about that. Yeah.
01:00:57.18
charliereese
Some people know how to build with steel and you could build a house out of any of those things. Some people not build with brick. Um, all of those things would build a house relatively well and the fastest way to build a house if you knew one of those skills wouldn't necessarily be to learn another one that might be slightly faster or cheaper.
01:00:58.16
Tom
Under.
01:01:35.86
charliereese
Would be to use the skill. You know to build it and get it done I guess where I'm saying there's a caveat is your house might one day need to turn into a skyscraper and there are things that you can use to quickly build a house that could also the same materials could be used to build the skyscraper. None example of that is concrete right? You could build an ugly house of concrete and if you happen to be good at building with concrete. Um, you could also build the skyscraper out of it. It would be very challenging and there would be practical reasons why you might not want to build the skyscraper out of wood one of which being fires.
01:02:42.10
Tom
Are.
01:02:54.40
charliereese
Um, and so yeah I guess what I'm saying is I feel on the backend that there are some frameworks that are very good at quickly getting a product out but are also quite good at scaling and it's been proven time and time again that successful companies have been able to use them and rails would be a very strong example of that. People are able to get products out really quickly with it but people have also scaled products to very very large like github and shopify and for a while Twitter um, although that was kind of relatively far back in the past now and they ended up moving away from rails and rails wasn't scalable back then and people were having more trouble. Um. Kind of like distributing computing on the cloud. So scalability of the actual framework in question mattered more It's harder to do that and now it's really easy to spin up like a None web servers. So the actual scalability of the framework isn't very important anymore. Um, but yeah I guess that's my caveat. I feel like we've gone on a tangent about this for so long now I've gone on a tangent about it for so long. But yeah I think I think it's one of those things where use what you know? um, is the short answer and then the long answer is like but if you were totally optimizing.
01:04:40.88
Tom
They are.
01:05:20.48
charliereese
Like maybe it makes sense to learn something else as long as your timef frame is really really long. Yeah I'm a stickler for the details done.
01:05:29.64
Tom
Um, fair enough. Um, and if we're talking about infra right? Just ah, the approach that I take is mostly what gets the job done as fast as possible mapping to what I know and then. Move on from that and like take take care of all the other things that need to be taken care of when you're trying to build out a bootstrap product. So if heroku works for you? Fantastic if you can link to Aws for like db and all that stuff and like you know aw else has all those things then also fantastic right? like.
01:06:17.62
charliereese
Sure yeah.
01:06:43.90
charliereese
Yeah, yeah.
01:06:41.68
Tom
I guess to mirror your infrastructure sort of framework I do I do right now because like I've just been like crazy about next like I do the versll and then link to Dbs that are like on digital ocean right? And for the most part. Yeah and like that solves the issue like.
01:07:00.68
charliereese
Yeah, yeah. Yeah I've been using for Celibate. It is quite good.
01:07:20.42
Tom
90 like that solves like the 90% of the things that I need to do with it right? and then when it comes to like backend stuff right? like you just spin stuff up on like docker and it's also just good to go as well right? So it's just whatever. It's just whatever kind of like deployment framework you kind of build as you're doing all these things and then you just kind of reuse them for each.
01:07:31.46
charliereese
Tip.
01:07:44.70
charliereese
Yeah.
01:08:05.36
charliereese
Yeah, although again I would say with the caveat. What's your timeline you know, um I have had I Guess yeah I Well I was going to say sorry go ahead.
01:07:59.30
Tom
New products that you're building.
01:08:20.16
Tom
It's always it's always short timeline because like getting getting Yeah sorry go I was just gonna say it's always your short timeline right? because you're trying to get to the next like sale or you're trying to get to the next picture.
01:08:44.20
charliereese
Well well it is at none right? like so so we've been kind of building small like hobby or side or whatever saas projects for how several years now right um. And I agree that all my timelines have been relatively short. But I've also been fortunate in the sense that maybe at a curiosity and maybe out of inefficiency who knows um I've tried different things I've tried different things I've tried deploying with docker like completely customized roll I ah. I've tried ecs for hosting my docker containers on aws I've tried heroku I've tried digitalocean done a little bit of Gcp stuff as your like everything. Um and I have found that some things work a little better than others. Maybe it's partially because of my background skills. Maybe it's partially preference. Maybe it's partially whatever. But again I do think it depends on your timeline and I think one of the really nice things about for me I really enjoy building saas products and when I'm doing it professionally? um. I'm doing it with much different motivations than when I'm doing it for myself. Even if it's a product that I'd like to be profitable if I'm doing it for myself. But what I'm doing it for myself part of what I'm trying to do is to find the most efficient way to build a new product. So I often try different things. Feel like I've slowly started zeroing in on what works very quickly. Um, and so while I agree that yes like Aws great solution. Whatever if you're a None or 2 man team. It's going to force you to think about things depending on what services you're using on aws it's going to force you to think about things that you might not need to think about. Um, one of which being ah like networking concerns and security groups where on other cloud infrastructure providers. The default settings might be secure enough and reasonable. Um, so so that would be an example of something that you could get started with right away and it's a great solution but it might not be the optimal one for a small team. Basically I'm just nit picking Tom and totally just if again, yeah, all solutions are pretty good. Um. Yeah.
01:13:20.88
Tom
That I mean that's what I have to say about this around front ends back end and infra.
01:13:33.16
charliereese
Yeah, yeah, and I guess all of my opinions are coming like kind of to bring it back to yesko no code low code. Um, if you're trying to code or sorry if you're trying to learn. Go yes code if you have a good reason to need to scale IE. you imminently need to scale better and things are breaking or things are growing really quickly and you need to make sure you can scale so that you don't shut down great. Do the esco thing um get super customized. You have the resources to do it. Awesome. Do the no code thing if you have to whether that is to save as much time as possible. There's already good solution off the shelf and so there's no benefit of doing yoga low code or yes code. Um, or you don't need to customize and then low code for everybody else I think is what you want to do? Um, and. I guess what we were just kind of arguing about is what's the best low codede solution and the best solution short term is kind of what you know and can implement and my feeling is that the best solution long term is one that you can efficiently most efficiently kind of. Implement none other solutions is that a fair is that a fair summary. Yeah big low code van over here. Love low code hate yes code at no good.
01:15:56.00
Tom
Cool. Yeah, yeah, that's for summary, you had a yeah you had a you had a little little mumble there when you were you you mixed up? Yes, and no and you're like yo code. What's yoco Charlie.
01:16:42.50
charliereese
Ah, ah yo code is yes and no code. Yeah, it's yes, no and low code. It's yo code. Yeah, it's this spectrum of codes that we were talking about yo code. Yeah.
01:16:55.56
Tom
It's all of it Yo character picking. Yeah.
01:17:20.20
charliereese
Like um, cool. Well hopefully people listening there was something useful in here for you for me if I had to give a takeaway if you're building something on a small team or for yourself and you're not hyper super scale like Netflix or Youtube. Find what you can off the shelf that works well for you. Hopefully there's something that's open source and free. Maybe there's something that's paid. That's reasonable to definitely do that. Save yourself the time engineering times expensive and also definitely do the lowcode thing if you need to do some customizing for the logic. Um. Reach for that. No code tool if it's already a perfect solution. You don't need to customize it or maybe you're not a software developer and you need to roll out a solution or you have 30 minutes and you need to roll out a solution reach for something off the shelf. They don't need to write a code for and then yeah, yes code I think I already covered that. But. Yeah, you need scale. You have good reason scale none Tom anything dad sweet all right cool. Well thank you for putting up with us while we were indefinitely on break kind of deciding what we wanted to do with the podcast.
01:19:07.66
Tom
Nothing else for me.
01:19:42.38
charliereese
Podcast just going to be more of the same but more focused on going from 0 to $10000 um rr going forward and we will catch you next week. See you guys.