Local first apps have some peculiar technical features, yes. I worked on CRDTs from maybe 2008 and in 2012 we made a collaborative editor. Local first, all data on the client, syncing by WebSocket. I remember when we debugged it we once had a problem resetting a document. It kept reappearing. The guilty was an iPad laying on the table face down. It synced all the data back. That was seriously different from e.g. Evernote that was losing data all the time. In our system, intentionally purging was really difficult.
Once we ran two weeks with a "poisoned" document that was crushing any server it was uploaded to. The user kept the tab open just working like nothing was happening. Then, we found the bug, but in theory we may have made the entire cluster restart all the time. Apart from electricity consumption, that would hardly change anything. The load and syncing time would be worse, but not by much.
With local-first, everything keeps working even without the server.
I purposely did not set up any infrastructure for my iOS app Reflect and made it local-first [0]. I did it so that I could make the app completely private but it’s come with the wonderful side effect that the product is easily horizontally scalable. The only overhead I have for more users is a busier discord channel and so far that community growth has felt very rewarding.
1. No network latency, you do not have to send anything across the atlantic.
2. Your get privacy.
3. Its free, you do not need to pay any SaaS business.
An additional would be, the scale being built-in. Every person has their own setup. One central agency doesn't have to take care of all.
By mkagenius 10 hours ago
And if you’re willing to pay $195 for the package and upgrades of $50 each time the OS breaks the app, we are good. And as long as this software doesn’t have to sync across mobile and desktop. And as long as this is single user software.
Then sure.
By ebiester 9 hours ago
Take Microsoft Office, if you just need Word, Excel, PowerPoint, Outlook and OneNote, that's the same price as a two year subscription to Office 365 / Microsoft 365 / CoPilot. That's potentially a good deal, depending on your needs.
Some people have been running Office 2003 on everything from Windows XP to Windows 10. Assuming they bought the license 22 years ago, that's pretty cheap, probably $15 per year. As a bonus, they've never had their workflow disrupted.
By mrweasel 9 hours ago
On the other hand, theres 17+ years of security updates missing.... and office macros are a known vector.
By privatelypublic 8 hours ago
You shouldn't upgrade the OS on a machine with important paid-for software which you are using.
By carlosjobim 7 hours ago
IMHO, this overlooks probably the biggest advantage of all: software you can buy once and run locally is predictable.
While the modern world of mobile devices and near-permanent fast and reliable connectivity has brought some real advantages, it has also brought the ability for software developers to ruthlessly exploit their users in ways no-one would have dreamt of 20 or 30 years ago. Often these are pitched as if they are for the user’s benefit — a UI “enhancement” here, an “improved” feature there, a bit of casual spying “to help us improve our software and share only with carefully selected partners”, a subscription model that “avoids the big up-front cost everyone used to pay” (or some questionable logic about “CAPEX vs OPEX” for business software), our startup has been bought by a competitor but all the customers who chose our product specifically to avoid that competitor’s inferior alternative have nothing to worry about because they have no ulterior motive and will continue developing it just the way we have so far.
The truth we all know but don’t want to talk about is that many of the modern trends in software have been widely adopted because they make things easier and/or more profitable for software developers at the direct expense of the user’s experience and/or bank account.
By Chris_Newton 7 hours ago
I think you missed the most important thing, more than any of those, if there is no service, then the service cannot delete or deny access to your data.
And related, if there is no service, then the service cannot fail to secure your data.
No possibility of a billing or login error, where you lose access to your stuff because they just think your're not current or valid when you are.
No possibility of losing access because the internet is down or your current location is geo blocked etc.
No possibility of your account being killed because they scanned the data and decided it contained child porn or pirated movies or software or cad files to make guns or hate speech etc.
Those overlap with free and privacy but the seperate point is not the money or the privacy but the fact that someone else can kill your stuff at any time without warning or recourse.
And someone else can lose your stuff, either directly by having their own servers broken into, or indirectly, by your login credentials getting leaked on your end.
By Brian_K_White 9 hours ago
> the service cannot delete or deny access to your data... the service cannot fail to secure your data.
You, on the other hand, are much more likely to do one or both of these things to yourself.
By jeffbee 9 hours ago
I will never decide that my kids doctor photos are child porn and delete all my own access to both my email and phone number to access my own bank and retirement accounts.
The fact that a hard drive can break and you can fail to have a backup is not remotely in the same class of problem of living at the whim of a service provider.
By Brian_K_White 7 hours ago
If we're tallying such things up, I've had those sorts of major issues with Trello, Google, Navionics, FitBit, FixD, and a number of other companies over the years. My local data on the other hand has had zero issues.
By hansvm 8 hours ago
I’ve seen this happen in a previous job where the IT team of a customer deleted years worth of financial records because they didn’t know about it when they cleaned up the server. Our CEO had to go and help them rebuild their books, at great cost to the customer!
By scott_w 8 hours ago
All of those advantages are also reason why businesses don’t adapt it…
By nicce 9 hours ago
It's open source, you "just" have to help write it!
By esafak 9 hours ago
This entire paradigm gets turned on its head with AI. I tried to do this with purely local compute, and it's a bad play. We don't have good edge compute yet.
1. A lot of good models require an amount of VRAM that is only present in data center GPUs.
2. For models which can run locally (Flux, etc.), you get dramatically different performance between top of line cards and older GPUs. Then you have to serve different models with different sampling techniques to different hardware classes.
3. GPU hardware is expensive and most consumers don't have GPUs. You'll severely limit your TAM if you require a GPU.
4. Mac support is horrible, which alienates half of your potential customers.
It's best to follow the Cursor model where the data center is a necessary evil and the local software is an adapter and visualizer of the local file system.
By echelon 9 hours ago
Define "good edge compute" in a way that doesn't have expectations set by server-based inference. I don't mean this to sound like a loaded request - we simply can't expect to perform the same operations at the same latency as cloud-based models.
These are two entirely separate paradigms. In many instances it is quite literally impossible to depend on models reachable by RF like in an ultra-low power forest mesh scenario for example.
By datameta 9 hours ago
We're in agreement that not all problem domains are amenable to data center compute. Those that don't have internet, etc.
But for consumer software that can be internet connected, data center GPU is dominating local edge compute. That's simply because the models are being designed to utilize a lot of VRAM.
By echelon 8 hours ago
> This entire paradigm gets turned on its head with AI. I tried to do this with purely local compute, and it's a bad play. We don't have good edge compute yet.
Your TV likely has a good enough CPU to run a decent model for home automation. And a game console most definitely does.
I'd love to see a protocol that would allow devices to upload a model to a computer and then let it sleep until a command it received. Current AI models are really self-contained, they don't need complicated infrastructure to run them.
By cyberax 8 hours ago
It's amazing to me how we called "box-product" now has a fancy new name "local-first". "Box product" is quite well understood. It has a lot of benefits, but also the business model is harder to get right comparing to cloud services. For opensource projects, that will not be a problem tho.
By 361994752 9 hours ago
I'm old enough to have bought software boxes in stores and I've never heard the term "box-product".
By plorkyeran 6 hours ago
Lot of young people don't know what a box product is. You're showing your age.
By 9283409232 9 hours ago
It reminds me of that joke where young people see a floppy disk for the first time and say, "Oh, you 3D printed the save icon!"
Might be reality very soon.
By 361994752 3 hours ago
Damn this hit me hard. If I ever hear this I know it's time to retire.
By 9283409232 2 hours ago
How dare people walk around showing their age. Disgusting!
By namuol 9 hours ago
Way to miss the point. They said box product is well understood but it isn't these days.
By 9283409232 8 hours ago
I'm 40 and I never heard that term.
By k__ 8 hours ago
Closest thing to true “serverless”: entire MVC app (Django/Rails/Laravel) in the browser with WASM and data persistence by SQLite over CDN.
All the server has to do then is serve binaries, all the business logic is in the client.
By myflash13 11 hours ago
What's WASM adding here? Without that you're just describing an ordinary SPA+CDN
By bcoates 10 hours ago
WASM adds the ability to run a local copy of SQLite (or even PostgreSQL) entirely in the browser.
By williamstein 9 hours ago
The ability to port existing apps to serverless. See for example Wordpress in WASM.
By myflash13 9 hours ago
Brilliant… but now you need to validate that the client did all their business logic correctly without tampering. That alone can be so complex it defeats the point.
By gjsman-1000 11 hours ago
No... you don't need that. Not for the overwhelmingly vast majority of cases. Let people use their own software. Tampering? Not my problem. Let people do it if they want.
By justinrubek 11 hours ago
The "overwhelmingly cast majority of cases" will be an employee of a larger company, a person/computer that cannot be trusted with arbitrary access to data and exceptions to business rules in code.
By Terr_ 10 hours ago
If it's a single-user app, you can only load data the user actually needs and is cleared to access. And/or lock down the device.
Multi-user app (and if we're talking about companies, it's multiple users by the very definition) where users are not trusted almost always needs either a central service with all the access controls, or a distributed equivalent of it (which is, indeed, very hard to implement). “Local-first” in those cases becomes less relevant, it’s more of a “on-premises/self-host” in this case.
But I think while end-user non-business software can be small compared to enterprise stuff, it is still a fairly big market with lots of opportunities.
By drdaeman 8 hours ago
Anything that involves sharing data with other people will run into issues around updating. If your API surface is a shipped sqlite db instead of an API call it's liable to be abused in so many ways.
By kevmo314 10 hours ago
Local-first doesn't mean local-only though, yeah? Isolate cloud usage to those collaborative features. If that's a huge ask, then your thingy probably isn't the kind of tool we're talking about localizing here!
By a_wild_dandan 9 hours ago
Anything that runs as a SaaS, or B2B, has that issue… which is the overwhelming majority of software.
Anything that requires sharing information with other users is also a pain in the neck, as you basically need to treat your internal logic like a proprietary, potentially hostile, file format.
By gjsman-1000 11 hours ago
There is a lot of SaaS that is essentially "for the buyer to the buyer", what I mean is that the software doesn't provide content to somebody else, or there is no incentive to serve malicious content (e. g. B2B). Why would tampering be relevant in those cases?
There are situations where it's relevant, but I don't think it's as many as you say
By Fire-Dragon-DoL 10 hours ago
The reason it's easier to scale local software is that it does not rely on cloud resources. As a result it's cheaper for a startup to distribute local first software since they don't need the infrastructure of a traditional cloud app. The problem is there is no business model for local first software like there is for subscriptions with SaaS. Traditional desktop apps were sold as single purchase items on CDs. That just doesn't work for local first software, since you probably just navigate to a website to get the software.
By danjl 10 hours ago
You can just charge subscriptions for local first software. No cloud is irrelevant. Only the value to the user matters.
By vimy 10 hours ago
Where do you check their subscription in order to cut off the service when they stop paying? One of the nice bits about local first is that there's no need for logins. Do you install security software in your local first app, which, typically, includes the code in a format that's fairly easy to bypass? Pirating desktop software was a big issue for companies. Are we going back to that horrible world?
By danjl 9 hours ago
A simple license key and monthly ping to license server (5 usd digital ocean) is enough.
Pirates is the cost of doing business. Just ignore them.
No need to make this too complicated.
By vimy 9 hours ago
So that kind of violates the principles of the local first software since you still need the cloud and license key in order to run the app. It also means more work for the developer, since they have no other reason to provide a server in most cases. It also means that they need to have logins which are not actually necessary for local first software and one of the benefits. Not so easy.
By danjl 8 hours ago
A license key is not a login. And a server like that is not a big deal.
Source: me, I do this way.
By vimy 7 hours ago
> Pirates is the cost of doing business. Just ignore them.
There are a lot of cracking groups that circumvent license servers on day one with software that have license servers.
I'm sure this is the reason that Adobe went to the cloud, Adobe couldn't ignore them as with other 'box software'.
By colesantiago 8 hours ago
> The problem is there is no business model for local first software like there is for subscriptions with SaaS.
I think this is too broad a stroke to paint with. There's local-first software that still connects to the cloud for additional features. Local-first can enable you to continue to work when offline, but the software can still be more useful when online.
By airstrike 10 hours ago
So your primary technical goal is to do local first, and you keep the cloud because it provides a business model that works? That feels very brittle. The way you're describing it, local first is an optional additional element for a cloud-based tool. I don't think that's the primary intent of the local first movement.
By danjl 10 hours ago
Developers can charge money, or subscriptions, for local apps as well. They can probably charge less, as they don't have a cloud provider to pay to host everything. This is pretty common with mobile apps.
By al_borland 9 hours ago
I remember Skype was local-first. I believe it was the only one commercially successful P2P project.
But over time and multiple hard-to-recover incidents they switched to cloud.
By deepsun 10 hours ago
Sure, you can try it charge subscriptions, but what are you actually charging for? You're not storing any of the customer's data or providing a service that they need, since the software should work entirely without your help. Fundamentally charging a subscription without having a centralized server is pretty tricky. Even the traditional desktop apps that transitioned to the cloud, like Photoshop or Maya, have really worked hard to beef up their cloud-based service to justify the subscription fee. A mismatch between your business model and your technical infrastructure is not going to stand up well over time. Don't get me wrong. I'm just trying to figure out what business model works for local-first software?
By danjl 10 hours ago
I think subscription models became associated with SaaS because cloud hype was at its peak around the time the first big corps were first migrating their products from perpetual licensing to susbscription, and just being on the cloud was seen as a selling point weighty enough to justify the price bump.
Now that cloud hype has died down, I don't see why subscription based would not be viable just because your product runs locally (assuming that all your competitors are already subscription based). ZBrush started selling local first subscriptions, so I guess we'll see soon enough whether that works out for them.
By Calavar 9 hours ago
Subscription models are associated with SaaS because you're selling a service. The service is typically storing your data or providing capabilities on the back end. With a local first app, the company is not paying for back-end resources, so there's a mismatch between the expectations of customers who are actually providing the resources on their own computer, and the desires of the company to make money.
By danjl 9 hours ago
> But over time and multiple hard-to-recover incidents they switched to cloud.
My understanding was that they switched to being centralized because phones couldn't run the decentralized version.
By thaumasiotes 9 hours ago
It depends on your customer base.The SCADA world is largely local, and believe me they have no trouble selling subscriptions.
By spauldo 9 hours ago
Their customers would probably prefer that there where no subscription though.
By mrweasel 9 hours ago
Probably, but you'd be surprised how little that matters.
By spauldo 7 hours ago
> Traditional desktop apps were sold as single purchase items on CDs. That just doesn't work for local first software, since you probably just navigate to a website to get the software.
Did I misunderstand this part? A lot of local software is sold as one time purchase downloads.
By carlosjobim 7 hours ago
> Traditional desktop apps were sold as single purchase items on CDs. That just doesn't work for local first software, since you probably just navigate to a website to get the software.
How does the reason you provide support the idea you provide it in support of? There are an infinite number of things that are sold as single purchases that you buy by just navigating to a website where you make the purchase.
There are an infinite number of things that are sold as single purchases on CDs that you buy by just navigating to a website where you make the purchase.
By thaumasiotes 9 hours ago
Local-first is the true server-less and your device is the real "edge".
So much truth to this post.
By samwillis 11 hours ago
Can we go back to native apps yet?
By iwontberude 10 hours ago
Local first is easier to scale technically.
Paid hosted software is easier to scale financially.
Without the latter, it's very hard to come up with the money to pay people to build, to support the market, etc.
By kylecordes 9 hours ago
I’m not sure it’s that clear…
Take an application like Slack and consider how to scale it local-first for a team with 1000 people.
And then consider how to coordinate deployment of new features and schema migrations…
Compare that with running a centralized server, which is going to be much easier.
By ccorcos 6 hours ago
Why local first and not a native application?
By kaycey2022 7 hours ago
"Lots of cloud providers like to brag about being able to scale with their users. I like to brag about not having to scale at all"
Bars.
By lazyhideousarom 9 hours ago
"Lots of cloud providers like to brag about being able to scale with their users. I like to brag about not having to scale at all."
Bars.
By lazyhideousarom 9 hours ago
While local first software might be easier to scale, the financial aspect remains unsolved.
I can see that Harper (The author's software) was acquired by Automattic so I assume that Automattic is paying the author to maintain Harper, effectively subsiding the maintenance costs to keep Harper free.
Not every local first (and open source) software has the opportunity to be supported by a big company.
By colesantiago 8 hours ago
An alternate article for this traffic spike is "PaaS is Easier to Scale". When the author relies on others to do the hosting and handle the scale the author doesn't have to worry about it. That's why he didn't need to be alerted. He's relying on others for that responsibility.
By charcircuit 10 hours ago
The authors rely on themselves, not other people. The developers of local first software take on the responsibility of making sure the app runs entirely on local resources. That's not the job of anybody else, and you don't need to pay for it like you do with cloud resources. This means it's cheaper for a startup to distribute local first software. The trick is that it's much harder to get paid for the app.
By danjl 10 hours ago
>not other people
The website is hosted by Automattic. The Firefox extention is hosted by Mozilla. The Chrome extention is hosted by Google. The Obsidian plugin is hosted by Obsidian. The VSCode extention is hosted by Microsoft. The source code is hosted by Github. The discord is hosted by Discord.
If you delegate your entire backend to other companies you won't be the one who has to worry about scaling.
By charcircuit 9 hours ago
With true local first software, there is no backend. Sometimes there's a static host that just delivers the code, but has no endpoints. Providing the code, or the extension, is not a service that the customer cares about. It's not for the customer. That's for the developer.
By danjl 9 hours ago
I'm talking about the developer having to worry about scaling and not the software. This article is about how the author got a sudden influx of traffic and did not have to worry about scaling anything to support it. That influx of traffic was to his website and then to the various distribution channels of his app. There is a reality that his website would fall over from the traffic and he would have to worry about scaling it, but in this reality he is paying someone else to handle scaling it for him.
If it wasn't local first software and he paid someone else for the backend, he also wouldn't have to worry about scaling the backend. My comment is pointing out that the dichotomy isn't between local first and non local first software, but between self hosting and not self hosting.
By gritzko 9 hours ago
By NoTranslationL 9 hours ago
By mkagenius 10 hours ago
By ebiester 9 hours ago
By mrweasel 9 hours ago
By privatelypublic 8 hours ago
By carlosjobim 7 hours ago
By Chris_Newton 7 hours ago
By Brian_K_White 9 hours ago
By jeffbee 9 hours ago
By Brian_K_White 7 hours ago
By hansvm 8 hours ago
By scott_w 8 hours ago
By nicce 9 hours ago
By esafak 9 hours ago
By echelon 9 hours ago
By datameta 9 hours ago
By echelon 8 hours ago
By cyberax 8 hours ago
By 361994752 9 hours ago
By plorkyeran 6 hours ago
By 9283409232 9 hours ago
By 361994752 3 hours ago
By 9283409232 2 hours ago
By namuol 9 hours ago
By 9283409232 8 hours ago
By k__ 8 hours ago
By myflash13 11 hours ago
By bcoates 10 hours ago
By williamstein 9 hours ago
By myflash13 9 hours ago
By gjsman-1000 11 hours ago
By justinrubek 11 hours ago
By Terr_ 10 hours ago
By drdaeman 8 hours ago
By kevmo314 10 hours ago
By a_wild_dandan 9 hours ago
By gjsman-1000 11 hours ago
By Fire-Dragon-DoL 10 hours ago
By danjl 10 hours ago
By vimy 10 hours ago
By danjl 9 hours ago
By vimy 9 hours ago
By danjl 8 hours ago
By vimy 7 hours ago
By colesantiago 8 hours ago
By airstrike 10 hours ago
By danjl 10 hours ago
By al_borland 9 hours ago
By deepsun 10 hours ago
By danjl 10 hours ago
By Calavar 9 hours ago
By danjl 9 hours ago
By thaumasiotes 9 hours ago
By spauldo 9 hours ago
By mrweasel 9 hours ago
By spauldo 7 hours ago
By carlosjobim 7 hours ago
By thaumasiotes 9 hours ago
By samwillis 11 hours ago
By iwontberude 10 hours ago
By kylecordes 9 hours ago
By ccorcos 6 hours ago
By kaycey2022 7 hours ago
By lazyhideousarom 9 hours ago
By lazyhideousarom 9 hours ago
By colesantiago 8 hours ago
By charcircuit 10 hours ago
By danjl 10 hours ago
By charcircuit 9 hours ago
By danjl 9 hours ago
By charcircuit 9 hours ago