Mike Hartington on the State of Mobile Development & Why CI/CD is Still So Hard

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

Get more video from Redmonk, Subscribe!

In this RedMonk conversation, Mike Hartington, Developer Advocate at Prisma, chats mobile development with Kate Holterhoff. They discuss mobile frameworks, the challenges of cross-platform development, and why CI/CD is so hard for deploying mobile apps.

Links

Transcript

Kate Holterhoff (00:12)
Hello and welcome to this RedMonk conversation. My name is Kate Holterhoff. I’m a senior analyst at RedMonk and today I have Mike Hardington, developer advocate at Prisma with me. He is alumni of NX and Ionic and Mike, I am so hype to have you here to talk mobile development with me.

Mike Hartington (00:30)
Yes, I I have been like in the mobile space for like so long it’s just like the thing I am like probably most known for and I got like I got the battle scars to prove That I have been there since almost the almost the very beginning at least very early on

Kate Holterhoff (00:50)
That’s wild. I had multiple folks tell me I should talk to you when I started realizing that even though I thought I understood mobile development, I actually didn’t. I was laboring under the illusion of competence, like usual, and realized my own sort of deficiencies here. Because I think as a front-end engineer, many of us are like, yeah.

We’re mobile front-end like these things can be kind of smooshed together and in a sense Maybe they are like in a labor sort of way a lot of teams kind of do that But when I was talking to one engineer He even mentioned that like there’s one person at his organization who understands mobile and everybody else just throws the mobile questions to this person and and this person is like, know if they leave they’re in big trouble because there’s like one person who really gets mobile and there’s all kinds of like

weird knowledge that goes into this. Like, it’s mysterious. So I’m excited to have you here, to pull back the veil Let’s go into your background first. So why are you the person to talk to about mobile stuff? What have you seen in the past?

Mike Hartington (02:00)
I I would say I am a person to talk to. There’s probably, like you said, there’s more people out there who have this knowledge just at their core. But my experience was I was a pretty early adopter of some mobile tech at many, many jobs ago where we were building cool trade show kind of portfolios that we wanted to bring out instead of bringing paper.

Kate Holterhoff (02:10)
You

Mike Hartington (02:30)
Our CEO at the time brought out an ipad and that was like the hot new fancy thing and so I built a bunch of stuff to like Make a mobile app Which at the time was kind of not a thing people did And this was like 2012 Right around then so yeah, so it’s been a while and ever since then I just have been in the mobile space kind of

Kate Holterhoff (02:35)
Mm-hmm.

wow, yeah.

Mike Hartington (02:58)
Keeping tabs on like what are the new features coming out in each like iOS and Android? What are the new tools and frameworks that can be used to? You know the back to your example like to simplify it so more people have more accessibility to develop a mobile app the cross-platform story, I think is probably the thing I’m most known for and Yeah, just kind of I have been in that space for so long. It’s just it’s a part of me It’s like who it’s kind of part of my identity for better or for worse

Kate Holterhoff (03:32)
Yeah. Okay. Okay. And so for folks who aren’t familiar, what is Ionic? Like, did you do there and like, what is their product?

Mike Hartington (03:42)
Yeah, so Ionic was a… started off as a UI framework for mobile apps, for cross-platform mobile apps. The idea was that you could build a mobile app using web technologies, HTML, CSS, and JavaScript. And there was this existing tool out there called PhoneGap or Cordova. It’s kind of like one’s the open source, one’s the commercial. We…

like that idea, but we’ve said the only downside is all the apps look like garbage that are coming out of this, making people think like, no real app could be built with this stuff because look at the apps that are out there. We built a framework and this collection of UI components that would enhance the bare bones apps that people were building. You would use our like…

scrolling containers, our buttons, we’d give you animations. We basically would sit down, record stuff on mobile devices, like how long does an animation last? How long does all this, does this interaction last inside of an app? We would nerd out about those details, re-implement them for the web, and then have that be like an open source project anyone could adopt into their code base if they want to go down that road.

Kate Holterhoff (05:04)
because now there are a few ways to do this. Ionic was sort of an early entrant into this. And there’s also these sort of buzzwordy things. And I’m hoping we can do some definitions as we go. okay, so you’ve got the multi-platform and you’ve got hybrid platforms. And Ionic, which one was it? Did they fit into that terminology? They’re a framework, so.

Mike Hartington (05:19)
Mm-hmm. Mm-hmm.

They were a framework and… Okay, so here’s the thing about that term, about like multi-platform hybrid framework. It’s kind of like two sides of the same coin, if you ask me. The thing that we kind of approached it was we were a web-first approach to building a Native mobile app. At the end of the day, yes, there were web assets in there, but there was still Native code for each platform that sat along with it.

Kate Holterhoff (05:34)
Okay.

Mike Hartington (05:58)
So if we were to think of it as a spectrum, like one would be full web, one would be full Native, we kind of sat like halfway in between where we had equal parts web and Native code. Now there are other platform and other tools out there that are further along towards the Native spectrum. And then there’s like pure web development where you’re just doing your typical front end engineering and not even considering like a Native mobile app.

Kate Holterhoff (05:59)
and so I, had historically thought of like PWAs or just web apps as being these, things that were, you we always did like mobile first when we were, creating front end designs. And so, you know, I was like, well, yeah, I’m a mobile developer, but no, no, no. In fact, it’s all about the hardware, right? It’s all about you got to learn some flutter, right? And

Mike Hartington (06:32)
Mm-hmm.

Yeah.

Kate Holterhoff (06:48)
I feel like this is top of mind now because there’s a lot of these frameworks coming out. not, know, the, the air in the room seems to be taking up with Expo. They’ve got their EAS thing, great. And then React Native was sort of the default, especially if you want to try to get some React developers. And then as you mentioned, you know, we got some Cordova in there, which is like, I don’t know, do you want to call it legacy? Is that cruel? I don’t know. But like Cordova has been around for a while. But did you hear about Kotlin Multiplatform, right? KMP. So.

Mike Hartington (07:08)
I think so.

Yes.

Kate Holterhoff (07:14)
yes. so now this i believe this is this year right so so now there’s even sort of more folks that are playing it here so i feel like there’s this momentum behind it i don’t know so yeah so you’re a little it’s two sides of the same coin we’re still getting some Native in there i don’t know

Mike Hartington (07:30)
Yeah, I think with the Kotlin multiplatform, like the fun thing, there’s also a Swift UI version for this where there’s like they’re trying to do Swift in Android now. There’s like this very interesting kind of goal that everyone was eventually going to get to where you have this collection of developers who know a particular skill set.

Kate Holterhoff (07:41)
goodness.

Mike Hartington (07:56)
This could be Java developers or Kotlin developers. This could be Objective-C Swift developers. And they all want to just ship something to their users. It could be shipping an iOS app. Okay, now how do I get this app to my Android users without going out and having to spend extra engineering effort into re-implementing everything?

hiring people, it’s time, money, it’s just a very slow and expensive process. How can I reduce that, like, that whole entire timeline so that way I can just take an app? I already have the core technologies, I just want to ship it out there. And so Kotlin Multiplatform is like the, like, very obvious thing that, cool, Kotlin has a very nice abstraction around building an app.

Let’s figure out how can we get Kotlin onto iOS? How can we get Swift and SwiftUI onto Android? So everyone’s trying to go for the same goal.

Kate Holterhoff (09:02)
Right, okay, so we’re all funneling ourselves towards having mobile. And so this is where I was like trying to figure out these gaps in my own ways of thinking about this problem. and let me explain what I mean there. So you think of like having one code base, and I know you worked at NX, so Monorepos, I guess we could bring this in here if we must. you know, so was like, well great, then you have like your React Native code and it can run.

Mike Hartington (09:19)
Mm-hmm.

Kate Holterhoff (09:29)
on the web, you know, on my desktop, and it can also run on all these mobile devices. But in fact, you actually have to use like special, frameworks to do this. So a lot of companies actually have two front end repos instead of just the one that I feel like was advertised. Like I hear React Native, I’m like, yeah, one code base, that’s all you have to change. But no, no, no, no. Okay, are we gonna get to that sp-

Mike Hartington (09:39)
Mm-hmm.

Kate Holterhoff (09:55)
this place because it sounds like there’s a few little experimental things that can do this, like I get the impression that this is not widely adopted. Like there are multiple. So I’m thinking of like Nicolas Gallagher’s React-native-web right, as like a way to bridge this. But in general, what I’m hearing on Hacker News and on Reddit is that you have the two front-end code bases. So anyways, are we gonna get to one? Is that even desirable?

Mike Hartington (10:09)
Mm-hmm.

I’d argue it’s desirable until it isn’t. It’s like every answer is like it depends because like if you have like iOS and Android are uniquely different platforms and the web is a uniquely different platform. So having everything just be look like a cookie cutter version of your app on all these platforms, it kind of misses out on the nuances and the

Kate Holterhoff (10:30)
Yeah, sure.

Mike Hartington (10:50)
the specifics of each of those platforms that you would want to take advantage of. Like, it makes sense to want to get as close as you can to that goal, but there’s still going to be some parts where that project might have… Okay, this is where we fork out for iOS, for Android, for the web, just because of this one very specific reason.

But you could still have it, it still exists in a code base and then you just share that special access to that art for certain teams.

Kate Holterhoff (11:19)
Right.

Yeah. Okay. Okay. So it depends. Fine. I guess. Okay.

Mike Hartington (11:31)
It’s like the very safe answer that no one ever loves and no one likes to get, but it’s true. It’s like… It’s a weird place to be working in.

Kate Holterhoff (11:38)
Yeah, okay. Okay.

Well, are monorepos the solution to this problem at all? I’ve heard that. I’ve heard presentations at Devnexus that actually pointed to there being ways to, leverage a monorepo in order to have this parity and avoid extra work.

Mike Hartington (12:00)
Yeah, I think a monorepos like a smart way to to do this I know during my time at NX I built a Java back end a React front end and then a Java Kotlin Android app that everything could just be consuming the same Endpoint and I’m sure if I had like put in the extra effort I could start playing around with like here’s my

frontend slash web and then here’s my frontend slash iOS and my frontend slash Android and I have everything compartmentalized into these very unique directories and parts of my monorepo and it just makes managing and development much simpler. There’s a trade-off though because now you’re taking on the ownership of how do I take all these separate pieces and assemble them to make sense.

Kate Holterhoff (12:47)
All right.

Mm-hmm, right. So you’re just trading off complexity. Well, I did find that illuminating because it does feel like a lot of these, at least the marketing terminology around these multi-platform and hybrid is that there is gonna be just one code base to rule them all. And I get the impression that that’s just not the case. And frankly, I talked to some more grumpy developers who said, no.

Enterprise, guess customer whatever you even use like React Native they would all have different code bases that You would only want to use like flutter or something for your mobile experience rather than your back end and your your actual Yeah desktop experience. I mean, what do you yeah? What do you say respond to that? What do you think? Is that not true?

Mike Hartington (13:30)
Hmm

I mean, one of the biggest companies in the world uses React Native all the time. Microsoft is like the… They are contributors to React Native for a reason. They use it for a lot of their… A lot of their products are just React Native by default because they have the… Velocity already baked in. Their teams are already up and running with it. They know what they’re doing and they know that it’s…

Kate Holterhoff (13:49)
Yeah.

Mm-hmm.

Mike Hartington (14:07)
makes sense for them. everyone who’s like any absolutist approach of this would never fly in this environment.

Kate Holterhoff (14:09)
Mm-hmm.

Mike Hartington (14:18)
It’s already there. It’s even if it’s not like Obvious, it doesn’t matter. It’s still there

Kate Holterhoff (14:24)
So, it’s already happening. And so, if I meet one of these, I wouldn’t say anti-React Native, but just like React Native doesn’t belong in standing up a serious app, I can point them to folks who’ve done it. But is there like a technical thing I can point to as well as like, like was there?

Mike Hartington (14:26)
Mm-hmm.

Mm-hmm.

Kate Holterhoff (14:41)
a moment in recent history where it became viable in ways that it hadn’t been in the past.

Mike Hartington (14:46)
probably say like React Native had their moment where Microsoft shipped a React Native bridge for building Windows apps and then Microsoft also shipped several Windows like starter apps as React Native apps instead of like pure native whatever code base is used for.

Kate Holterhoff (15:10)
Yeah.

Mike Hartington (15:11)
Windows itself, they ship like part of those apps as React Native apps and it’s like Microsoft can do it There’s kind of no reason not to like hold off on it I mean there are some reasons to hold off but the Microsoft is like a leader where you can say hey technically this shipped and it’s great and people don’t care You should reevaluate your opinion

Kate Holterhoff (15:17)
That was it. Yeah, okay.

Okay.

Yeah. I wonder if it’s that a lot of times these architectural decisions get made by folks in the back end and maybe they’re not having the right folks at the table to push back and say, no, actually, React Native is, you know, it can scale, can do all the things that we want and maybe they’re just not being listened to. Because, yeah, like I said, it it tended to be folks that’s mobile is not their specialty. And, you know, like I.

found on my own is I feel like the mobile developers, there aren’t like huge thought leaders in the mobile development space. There’s vendors. But I have had a hard time trying to find like who to follow to keep up on the latest and greatest in mobile. I mean, I don’t know, can you can you help me with that, Mike? Like, who do you how do you keep up on what’s going on in mobile?

Mike Hartington (16:10)
Mm-hmm.

So I’ve gotten a pretty good group of people that I follow. So for like the pure native side of the house, there’s a gentleman from Firebase. Peter Friese. He works at the Firebase team, but he’s also been like a big Swift and native mobile like Apple ecosystem kind of advocate for a while. And so like he has a whole series on how do you build something in Swift and keep up to date with the Swift UI changes and all everything going on in Swift mobile.

Kate Holterhoff (16:37)
Okay.

Okay.

Mike Hartington (17:03)
Check out his stuff and I follow him on Bluesky React Native there is Brent Vante who works at Expo I’ve gotten to meet him a few times and he is always sharing the cool things that they’re doing in that space Android is a little bit of a different one. I just kind of watched a Google I/O keynotes and Hopefully hopefully I’m at least familiar enough with what’s going on in there to be like, that’s cool

Kate Holterhoff (17:06)
Great.

Yeah.

Yeah.

Mike Hartington (17:33)
I guess I gotta go talk to a few people and see what’s the big deal about this.

Kate Holterhoff (17:33)
Okay.

Okay, And I’m glad you brought up Firebase because they were definitely on my list of things that we need to chat about. The reason for that is, I mean, at a high level.

Mike Hartington (17:43)
Yes.

Kate Holterhoff (17:47)
Who is making money on mobile, right? You know, I’ve had the conversation with folks like Alex Russell about the app store and Apple’s policies and that. And then I feel like that’s almost like a separate conversation. I think what I’m interested in is like, when I either buy or just download an app, it’s basically a zip file, then like is it?

Mike Hartington (18:11)
Mm-hmm.

Kate Holterhoff (18:14)
It doesn’t seem like it would be hosting, like that mobile developers are making money through hosting. It seems like it’s something like advertising or like, when you purchase something, like that’s where folks are making money. So, and when I look into this, just seems like Firebase is the default here, but then I suspect it’s going to be like hosting providers, the hyperscalers, whoever.

That that’s who is making the money because they they have the servers, right? So I guess my question to you is like is there some sort of market that I’m missing here

Mike Hartington (18:41)
Mm-hmm.

My my experience actually shipping apps to the app store and making money is almost non-existent So I never went into it trying to make money. I had a stable job Most people yeah, it’s a lot of it is ad driven I think like when back when I was at ionic one of the most common Request and common pieces of like tutorial information people requested was like

Kate Holterhoff (19:04)
ad driven.

Mike Hartington (19:14)
How can I implement ads in my app? Because I want to make money somehow. And especially if you’re trying to deploy to iOS and you have to pay the $100 Apple fee for your certificate. It’s not cheap. And not to mention the hardware. yeah, ads tend to be the biggest thing that people use. And that’s like a revenue split because you try to get some. And then you have to also

Kate Holterhoff (19:28)
Yep.

Mm-hmm.

Mike Hartington (19:44)
pay the piper and pay the people who provide the ads and there’s that revenue split and if you have in-app purchases there’s a revenue split for reasons. Yeah, the real people who make the money aren’t the ones doing the actual hosting for like you have assets that you’re downloading or storing. You’re gonna have to pay a file storing so your AWSs will definitely collect some money from there.

Kate Holterhoff (20:13)
Mm-hmm.

Mike Hartington (20:14)
It’s just the kind of nature, like the success margin for apps that get published and become these billion dollar companies or these success stories that you hear about, like you kind of used to hear about more often very early on in mobile, those don’t happen too much anymore. And how…

Kate Holterhoff (20:33)
Yeah.

Mike Hartington (20:35)
How people sustain mobile app development is kind of a, it’s kind of one of those things I never got a good answer for.

Kate Holterhoff (20:43)
Interesting. OK, well, clearly, I’m with you. I am, Well, let’s shift it slightly and talk about dev tools around mobile development. I talk with folks who do observability around mobile development, some of them in the OTel community, OpenTelemetry. But it seems like there should be more. And so I guess a way of framing this question for you would be,

Mike Hartington (20:52)
Mm-hmm.

Kate Holterhoff (21:12)
Who would you say is the Vercel of mobile development?

Mike Hartington (21:19)
That’s a good one It is because there really isn’t somebody out there that is doing The things that Vercel does at that scale It is a very weird place like Sentry was like a big player in there for for a bit cuz there were they had like the best like kind of SDKs for error reporting, but there was no like larger

Kate Holterhoff (21:20)
Isn’t it? Isn’t it a good question?

Mike Hartington (21:46)
place that you go to and that’s kind of the downfall of like mobile as a a as like a platform is that you are piecing all these things together and duct tape and hoping that things really work and communicate together and okay i gotta do some like i’m gonna send this off to one service to then massage it to go into this other service and then hopefully things work out just fine

Kate Holterhoff (22:00)
Yeah!

Right, right. Yeah, so duct tape and bailing wire right now. So it does seem like there’s an opportunity there. And I have heard that Expo is kind of filling that gap. It seems to me that maybe Firebase is too. But it seems to come down to the CI/CD pipeline, that just the deploying part of these mobile apps is the challenge. And

Mike Hartington (22:38)
Mm-hmm.

Kate Holterhoff (22:39)
until the App Store situation is, I don’t know what, evolves, I guess, in the sense of not having to wait for approval and pay this money and whatever, that we won’t have a real easy deployment, one-click Heroku-style, beautiful deployment experience, right? That this just, it isn’t possible in the current…

set up, which is too bad, right? I mean, that we’ve got this gatekeeping around devices that are in our pockets, right? I mean, it’s just a computer. So, I don’t know. Okay. I didn’t want to lead you, though, by bringing up the CI/CD thing, because Would you agree that that’s the hard part? Okay.

Mike Hartington (23:20)
I it is the hard part, because everyone would get to the point of like, I have this app, I’m testing it locally, now what? It was so hard that like even at Ionic, we built a whole commercial like leg of the company to solve that problem and to get deployments simplified to the fact where it’s like, you push to git, we trigger a build and like give you an APK or that package format.

Kate Holterhoff (23:28)
Yeah.

wow.

Mike Hartington (23:49)
And then you can test it, deploy it, or upload it to whatever store. And that was something that everybody in our ecosystem was thrilled to have. I know Expo has their EAS, which is a huge win for the React Native folks. Because it’s just as hard with React Native than it would have been with Ionic. How do I actually take my thing and ship it out to people? Native Android, Native iOS.

It’s all just a challenge and I think that’s kind of the I think they kind of intentionally keep it that way. This is me kind of a tinfoil hat moment like There was for a small period of time and made in by small period time I think like maybe two or three days where Apple put out like this Apple’s App Store standards kind of paper Or webpage and

Kate Holterhoff (24:21)
You

Ooh. Ooh. Go on. I’m here for it.

Mike Hartington (24:48)
There was some verbiage in there that was like, this isn’t amateur hour. This is the pros only. And it was very much a, don’t come to the app store if you’re not a professional who’s ready to do the work. They removed that language and kind of reverted back to a more friendly, open, kind of explanation, but it never really settled in my mind that they were willing to make

the store and that whole shipping experience better. It just kind of like, you can take the words back and remove them from the site, but the sentiment is still there. And the lived experience of having to ship an app through your platform is that it is intentionally high, probably to reduce people who would ship less than ideal apps. But it seems very unfair, especially as someone who has no control of that.

Kate Holterhoff (25:48)
Yeah, whoa. And so, yeah, no. Keep it to the side. Don’t throw it away. Have you had that experience with Android? Is there just this huge philosophy difference between these two predominant app stores?

Mike Hartington (25:50)
Tin foil hat off.

Android is a little bit easier to deploy to mostly because the hardware requirements that you have If you wanted to ship something to iOS you needed a Mac if you want to ship something to Android or to the Google Play Store You didn’t need a specific piece of hardware. You could do Windows Mac Linux Now the the counterpoint is that anyone could ship an app to Google Play

and Google Play does not have the reputation for having stellar quality apps. there is kind of a… Apple does have a good reason for why they maintain such high quality and excellence and they have a review process. Google Play is a little bit more lenient, but you get some stuff that could be sketchy or just questionable at best.

Kate Holterhoff (26:34)
Mm-hmm.

Okay.

Okay.

Got it, got it, got it. OK. Is it also less expensive to deploy on Android? Wow.

Mike Hartington (27:07)
significantly. I think the fee was like 20 bucks back in the day. I remember paying

20 bucks for an Android development license, like Google Play publishing license or whatever it was. It might have gone up, but it was in the grand scheme of things I could justify that as like, oh I guess I just won’t go out to lunch one day.

Kate Holterhoff (27:29)
Yeah, right. Yeah. No, that’s fine. We can talk historically. Folks can check up. And frankly, I’m sure it’s one of those things that changes frequently. yeah. But in your experience, historically cheaper. OK. So it seems like there is we talked about how things are moving. We talked about the Kotlin platform. So I’m interested in some of these new open source products. So of course, we’ve got Expo. We’ve talked about them a little bit.

But we’ve also got now Snapchat’s Valdi and ByteDance’s Lynx. Have you messed around with those at all?

Mike Hartington (28:01)
Yeah. Mm-hmm. I honestly, I just saw the Snapchat one. I, yeah, I looked at it and I was like, it definitely seems like it’s, it came out of their use from React Native. It’s a clear, obvious, like, this is where we came from. It’s interesting to me because everybody,

Kate Holterhoff (28:09)
yeah. Well just came out. That makes sense. Okay.

Mike Hartington (28:30)
is trying to rebuild the same thing and I don’t know I don’t have I don’t know the actual term for it I like to call it the pink wheel syndrome like you’re not reinventing the wheel you just made the wheel again but you painted it a different color and try to pass it off as being this brand new revolutionary thing

Kate Holterhoff (28:50)
Mm-hmm. That is really good. I’m gonna steal that. That’s a very good explanation. Pink wheel. Okay. All right, so they pink wheeled it. All right, and links, have you seen that?

Mike Hartington (29:02)
It looks, yeah, I saw that the ByteDance one, kind of came, didn’t, didn’t surprise me too much. Um, cause you know, Tik Tok such a huge, huge app. They have probably so much, uh, technical like experience in there that they were like, makes sense for us to kind of take something and do it on our own. And I can only imagine the effort they put into it. It looks cool. I haven’t built anything with it. I haven’t.

Kate Holterhoff (29:14)
yeah. Wow.

Mm-hmm.

Mike Hartington (29:32)
found the need, but definitely something to keep a keep kind of tabs on and just be like, cool, what is this? How is this evolving? What are they doing different? What are they trying to implement that you couldn’t get from, say, your React Natives?

Kate Holterhoff (29:33)
Okay. Yeah. Okay. And in the spirit of I’ve had a hard time finding thought leaders in this and just kind of find out about it through my front end channels, are there others I’m missing that have been open sourced recently?

Mike Hartington (29:59)
Would you consider Flutter open source? It’s not recent, but it out there.

Kate Holterhoff (30:03)
I mean, no, it’s not recent at all. Absolutely. And that’s the thing is when I have looked into like Expo, you really see this, I don’t know if you want to call it like tribalism between the Flutter folks and the React Native folks. And they will not meet. they, I feel like like Hacker News is populated with the Flutter people and they love Flutter and they will hear no, you know, they won’t change. So I, I…

Because of that, though, it’s hard to find, like, really compelling comparisons between them. They just seem to be sort of comparable. I mean, if anything, they’re kind of like, well, it’s a bunch of front-end engineers working on React Native. The real developers are Flutter developers. I don’t know. Maybe I’m projecting here. But yeah, I only excluded them because of, they just don’t seem to be like…

the new sort of like vendor driven approaches. but I mean, you tell me, yeah, why do you include them in this list here? Like, you just feel like it’s comparable. And so if we’re thinking of different options, Flutter’s gotta be in there.

Mike Hartington (30:58)
Yeah. It’s comparable, yeah, like I would definitely include it. There’s also like a community fork of Flutter that started probably 2023, I would say, where they were like, we wanted to move faster. Google isn’t doing the Flutter project like any other

Kate Holterhoff (31:24)
Okay.

Mike Hartington (31:38)
service by staying slow. Yeah, I just had to look it up. It’s called Flock.

Kate Holterhoff (31:41)
Yeah.

FLOCK. F-L-O-C-K?

Mike Hartington (31:47)
Yeah, basically just like the Flock flutter. Yeah Bird metaphors everywhere But yeah, they took it they’re like this stuff is cool and we don’t want to see it go away and we want to make sure that we’re keeping up to date with what the What people are doing? with flutter, so we want to

Kate Holterhoff (31:50)
flock of birds. Yeah, got it. Okay, I got a bird, bird metaphors just everywhere. Okay.

Mike Hartington (32:14)
We’re gonna fork it and control it and kind of bring it back to the community. And it seems pretty cool because it’s like a drop-in replacement and it swaps out some of the underlying libraries, but it just looks cool. I’m pretty excited to see how far it goes, but it has been a few years.

Kate Holterhoff (32:15)
Control it.

Yeah.

Okay. Yeah. I mean, I wonder, do you think that the Flutter community was concerned about like a killed by Google situation that like Google, you know?

Mike Hartington (32:43)
I that’s the fear of every Google project. say with like, I’m involved in some of Google’s projects, like as a ambassador, but I don’t think it’s a secret that there is a high turnover rate to some of Google’s projects. I would say it’s doing Google what it’s doing for Google what they want it to do. And I don’t think there would be a reason why they would just come out and be like, we’re killing it.

Kate Holterhoff (32:47)
Well.

huh.

Yeah.

Mike Hartington (33:12)
Google’s a huge Java Dart. Like they do use that all the time and make sense for them. So it serves their purposes.

Kate Holterhoff (33:18)
Mm-hmm.

All right, Mike, AI. How is AI changing mobile development?

Mike Hartington (33:23)
I think So as as an engineer AI would is probably gonna make mobile development far easier in some some ways I built after I left Ionic one of my first things I want to do is build a mobile app for Android For iOS and I did use AI to help like as an assistant and not as like a vibe code

Kate Holterhoff (33:57)
Okay.

Mike Hartington (33:58)
I will say the AI got me pretty far until it broke and then I actually did have to learn. So it will help you get to that first kind of checkpoint and then you kind of have to like, now I need to refine how am I asking it? How am I using these tools to further focus everything? As an app builder who would want to integrate AI into their app, it’s twofold. I think it makes sense.

Kate Holterhoff (34:24)
Mm-hmm.

Mike Hartington (34:28)
for like a lot of the Like fun one that I had today was I want to integrate an AI imaging Model into my app so I could take a picture of something and turn it into like an 8-bit 32-bit like an old retro gaming kind of icon I was like cool. I don’t have to go ahead and do that myself. I can just go reach out to like open AI or Anthropic or any of the AI providers and be like

Kate Holterhoff (34:47)
cool.

Mike Hartington (34:58)
There we go. Here’s the image. Do the thing for me.

Kate Holterhoff (35:01)
Yeah. neat. yeah, that’s really cool. So what does that look like typically? Like APIs or there like iframe experiences or?

Mike Hartington (35:10)
Mostly a API driven I know for Anthropic the stuff I kind of I had worked with before it was just install this package you’re making requests against their own a their own API and you’re picking out which Which model and which kind of like Tier of tools you want to use and then you get the response back and it’s kind of like a live streaming of

Kate Holterhoff (35:13)
Yeah.

Mike Hartington (35:38)
Requests that come back from the server depending on how you set it up So for like this hypothetical image tool I would be basically taking a photo turning that into just random blobs of data sending it out to my AI Provider of choice and then hoping a response comes back and it can process it and then it’s playing it to my users It’d be a very like REST

Kate Holterhoff (35:42)
Mm-hmm.

Mm-hmm.

Mike Hartington (36:05)
similar to like a restful experience that devs have been using for years.

Kate Holterhoff (36:09)
Yeah. Right. They make it easy. Yeah. I know the OpenAI API has been around for several years now, and it always seemed like the default. But I do feel like they have some movement now. I mean, I’ve been thinking of GitHub wanting to make their HQ into a platform for leveraging agents. And so I do feel like there’s this move towards, where are we going to be?

Mike Hartington (36:28)
Mm-hmm.

Kate Holterhoff (36:33)
I don’t know, prompting our agents and running them and all of these things. And again, with mobile, it’s like, compute, is it happen, you know, if you’ve got the, say I’ve got Claude on my phone, well, that is using the edge to reach their servers. But increasingly with these smaller models, maybe it’ll be running on my phone at some point, or maybe it is already. I mean, I’m sure somebody’s done it. So just, feel like maybe there’s some movement there. But also you mentioned data, like.

You know, you’re in the data industry these days. Is there something that like either AI or just like the mobile space in general how are they thinking about where this information is stored? Is this just like local first, you know, trying to keep things on the device or, you know, are there preferred like databases for app developers who, want their Native apps running on their phones?

Mike Hartington (37:24)
I know, kind of bias. I’m pretty partial to Prisma. We do have some pretty good stuff, but I think as far as where people are storing all this data and everything, it’s kind of anyone’s guess. Apple for their credit tries to encourage people to do on-device first for privacy and security reasons. Android, not so much. There is on-device storage, but…

Kate Holterhoff (37:27)
Absolutely. Sure.

Yeah.

Okay. Makes sense.

Mm-hmm.

Mike Hartington (37:54)
you’re pretty much encouraged to push it to a Google-owned kind of data center. So there’s some possibly privacy-focused approaches that people want, should, and really have to take into consideration, especially with AI, as we’re seeing so many AI companies not respect user privacy, which is… it’s not really cool.

Kate Holterhoff (38:00)
Yeah.

Not cool at all.

Mike Hartington (38:24)
And in some places it’s illegal like everywhere except the US it feels like

Kate Holterhoff (38:31)
Yikes. Another conversation for another time, I’m afraid. But yes, indeed. OK. Yeah, I know the data thing, just at the broadest level, just seems to be, again, that issue of the place

Mike Hartington (38:36)
Huh.

Kate Holterhoff (38:49)
are running their compute, on the hyperscalers, on Firebase, whatever, is going to be the same for mobile as it is for these desktop the architecting is where things maybe change a little bit. And the whole reason I ask is because, yeah, as someone who is interested in the DevTools space, I’m just looking for how is mobile moving the needle on DevTools? How is it?

changing what we should pay attention to. Are there opportunities where maybe there’s a gap, like a need for new dev tools? And it just seems to come down not to the database, but to the CI/CD every time. Every time.

Mike Hartington (39:27)
Every time it’s still the hardest thing to solve and it’s still the biggest like time crunch in any Shipping because I said like everything can be reused if you’re building If you’re building a full company web infrastructure a lot of those resources are shared CI/CD is the only thing that can’t be shared for whatever reason

Kate Holterhoff (39:55)
Wow. Well, you have confirmed many of my suppositions. You have expanded them in other ways. I’ve learned a lot. Okay, Mike, I appreciate this. so for folks who want to continue hearing the exciting, prognostications that you have on this industry, where do you direct folks? Do you have a social media account? Should they find you on LinkedIn, email, blog? What do you do?

Mike Hartington (40:17)
Yeah, I am mostly active on Bluesky these days @mhartington.io basically on there I do I do occasionally blog I think I’m up to the one blog post a year and I try to make it a really good one Yeah, just Just for my own my own kind of thought process. I’ll be at mostly mostly mostly Bluesky these days

Kate Holterhoff (40:31)
nice. Ooh.

Yeah, you polish it for a year.

Mike Hartington (40:45)
and occasionally LinkedIn

Kate Holterhoff (40:45)
Okay. All right. Well, let’s wrap up then. I have really enjoyed speaking with you, Mike. Again, my name is Kate Holterhoff. I’m a senior analyst at RedMonk. If you enjoyed this conversation, please like, subscribe, and review the MonkCast on your podcast platform of choice. If you’re watching us on RedMonk’s YouTube channel, please like, subscribe, and engage with us in the comments.

More in this series

Conversations (106)