Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why haven't local-first apps become popular?

marcobambini.substack.com

Post reply on HN
Print view View original

171–180 of 494 posts

Re: Why haven't local-first apps become popular?

#171

Post by WorldMaker »

robenkleene wrote: post #93
jordanb wrote: post #85

Earlier quoted context omitted.

I feel like git set back mainstream acceptance of copy-and-merge workflows possibly forever. The merge workflow is not inherently complicated or convoluted. It's just that git is. When dvcses came out there were three contendors: darcs, mercurial and git. I evaluated all three and found darcs was the most intuitive but it was very slow. Git was a confused mess, and hg was a great compromise between fast and having a…

> The merge workflow is not inherently complicated or convoluted. It's just that git is. What makes merging in git complicated? And what's better about darcs and mercurial? (PS Not disagreeing just curious, I've worked in Mercurial and git and personally I've never noticed a difference, but that doesn't mean there isn't one.)

Darcs is a special case because it coevolved a predecessor/fork/alternative to CRDTs [0] (called "Patch Theory"). Darcs was slow because darcs supported a lot of auto-merging operations git or mercurial can't because they don't have the data structures for it. Darcs had a lot of smarts in its patch-oriented data structures, but sadly a lot of those smarts in worst cases (which were too common) led to exponential blowouts in performance. The lovely thing was that often when Darcs came out of that slow down it had a great, smart answer. But a lot of people's source control workflows don't have time to wait on their source control system to reason through an O(n ^ 2) or worse O(n ^ n) problem space. To find a CRDT-like "no conflict" solution or even a minimal conflict that is a smaller diff than a cheap three-way diff.

[0] Where CRDTs spent most of a couple of decades shooting for the stars and assuming "Conflict-Free" was manifest destiny/fate rather than a dream in a cruel pragmatic world of conflicts, Darcs was built for source control so knew emphatically that conflicts weren't avoidable. We're finally at the point where CRDTs are starting to take seriously that conflicts are unavoidable in real life data and trying new pragmatic approaches to "Conflict-Infrequent" rather that "Conflict-Free".

Re: Why haven't local-first apps become popular?

#172

Post by arrowtrench »

candiddevmike wrote: post #11

IMO, offline read-only is enough of a compromise. How many times are users truly offline, AND want to be able to edit at that time (and deal with the potential conflicts, which by the nature of the operation, won't have a good UX)?

Offline read-only would already be a great feature for me, knowing that my data is always close to machine. But, my guess is that this isn't enough of a killer feature for most people to care.

Re: Why haven't local-first apps become popular?

#173

Post by WorldMaker »

0xffff2 wrote: post #133

I feel like I'm taking crazy pills. How on Earth could anyone consider the example in #2 "conflict-free"? You haven't removed the conflict, you're just ignored it! Anything can be conflict free in that case. Obviously not every problem will have such an obvious right answer, but given the example the author chose, I don't see how you could accept any solution that doesn't produce "100" as a correct result.

I also think this is a place where CRDTs in general got stuck on the name "Conflict-Free" for way too long assuming it was fate that if they worked hard enough they find the magic data structures to eliminate conflicts altogether but real life data is a lot more more complicated than that and real life expectations of data semantics that a data type itself can't encode. I think we are just now getting to the point of seeing CRDT libraries understand some conflicts happen, and some conflicts still need to bubble up to a more complex semantic model or even/especially a user. I don't think there are any CRDT libraries that are strong for that yet, but the work seems to starting into those next steps at least.

Re: Why haven't local-first apps become popular?

#174

Post by threatofrain »

marginalia_nu wrote: post #24
lordnacho wrote: post #12

Local-first was the first kind of app. Way up into the 2000s, you'd use your local excel/word/etc, and the sync mechanism was calling your file annual_accounts_final_v3_amend_v5_final(3).xls But also nowadays you want to have information from other computers. Everything from shared calendars to the weather, or a social media entry. There's so much more you can do with internet access, you need to be able to access re…

> There's no easy way to keep sync, either. Look at CAP theorem. You can decide which leg you can do without, but you can't solve the distributed computing "problem". Best is just be aware of what tradeoff you're making. Git has largely solved asynchronous decentralized collaboration, but it requires file formats that are ideally as human understandable as machine-readable, or at least diffable/mergable in a way wher…

Gits approach is to make people solve it with organic intelligence.

Re: Why haven't local-first apps become popular?

#175

Post by ktosobcy »

Anonyneko wrote: post #125
ktosobcy wrote: post #53

I try to use those as much as possible. We ended up in this situation mostly because of the Google push to make everything an "web-app" with web-technologies instead of protocols... Anyone uses IMAP email? Works just fine (save for IMAP design but that's another story). Same with CalDAV. For stydy I use Anki and it has brilliant sync (it can even automatically/automagaically merge study changes when I study some item…

With Anki the sync works perfectly until it doesn't and the app asks you to choose the copy you want to keep. Thankfully it doesn't happen often and hasn't caused any trouble for me, but it (seemingly) has no options for manual conflict resolution.

Hmm... it requires full sync if you edit the fileds of template.

In the past it required full sync due to conflict but in the last ~2 years (maybe) I haven't run into such probelm (I guess periodic sync in the background eliminating to much of a backlock probably helps)

Re: Why haven't local-first apps become popular?

#176

Post by kelvinjps »

I think just doing files and then syncing the files with another solution it's okay.

For example my local apps: Synthing syncs the files between my computer and phone for Note taking(just markdown and org files) obsidian on my phone, emacs/vim on my PC. Todo and reminders: org mode in emacs (desktop) orgzly on mobile. Password manager: KeePassxc desktop, keepassdx mobile. Calendar: just sync the ics file. Photos: just sync files use the default app. I can continue with this with every app, I don't know why people overcomplicate things

Re: Why haven't local-first apps become popular?

#177

Post by reclusive-sky »

I've been working on a local-first distributed app and found it difficult to design something that is easy to install and maintain. Important prerequisites and user-managed dependencies (i.e. Redis, Docker) are nonstarters for most potential users. It seems cloud hosting is always the end-state when trying to optimize primarily for less-technical users. FWIW I also ended up arriving at CRDTs as the solution for my app.

Re: Why haven't local-first apps become popular?

#178

Post by arrowtrench »

SilverbeardUnix wrote: post #64

1. People don't work alone and need to collaborate 2. People value convenience over privacy and security 3. Cloud is easy.

Is local-first bad / more difficult for collaboration because of conflict resolution? (E.g., two users edit the same document when they're offline and then, during syncing, they find that their versions diverge too much for them to be merged cleanly.) If so, isn't it possible to mark certain assets as "undivergable" which would effectively mean that the program would act like a traditional cloud-first type of app for that specific asset? This middle ground could introduce too much complexity and some inconsistency, but it could prove useful in certain cases.

Re: Why haven't local-first apps become popular?

#179

Post by R_Spaghetti »

kelvinjps wrote: post #176

I think just doing files and then syncing the files with another solution it's okay. For example my local apps: Synthing syncs the files between my computer and phone for Note taking(just markdown and org files) obsidian on my phone, emacs/vim on my PC. Todo and reminders: org mode in emacs (desktop) orgzly on mobile. Password manager: KeePassxc desktop, keepassdx mobile. Calendar: just sync the ics file. Photos: jus…

This is the reason why I'm reading this site - this is such a brilliant idea, super simple and without any vendor lockin at all.

Re: Why haven't local-first apps become popular?

#180

Post by carlosjobim »

poisonborz wrote: post #162
carlosjobim wrote: post #155

Earlier quoted context omitted.

Both what? What problem are you solving?

I don't really get "local apps with sync". Sync to what, each other? In most cases there needs to be an always-on central authority (where backup also happens). My point is that "some device sitting alone in a corner with software and data stored on it" model is outdated (this is the "problem", I hope the "why" is obvious). But "local" should live on, just in the form of users owning most of their infrastructure.

Sync through iCloud or through the app manufacturers servers. Most apps I use are real apps that live on my local device. Then some of them sync through iCloud if I want to use sync. This is something I as the user can enable or disable in the app, and it costs me no subscription fee.

I think that this is much superior to using fake apps that live on some web server. It comes with great drawbacks even when it is on the web server of the app manufacturer. Ten times more pain if I have to host it myself.

Post reply on HN