I think that many existing apps with huge userbases will gradually lose users as the models become better and better. Their biggest advantage is that people don't like change, and thus having to e.g. export data from some tools etc. seems to be a hassle not worth $5 a month. But as the models get better and the quality of the output will match the quality of the established SaaS but tailor the whole thing to a single user with the ability to make any change they can imagine within minutes, and perhaps deploy to Hetzner and whatnot where they could host all of those apps for a single $5 instead, the exodus will accelerate.
On the other hand, new products will have a much harder time to gather a big userbase. Whenever I'll see a launch of a SaaS asking for $$$, the first question I'll ask myself will be how long it will take LLM to recreate it. And for most cases, I imagine that the time it will take to get 80% of what they have is a few vibe coding sessions (as most newcomers will probably have used LLM themselves to code it up).
I do think you're vastly overestimating people's ability to write software, even with LLMs, and use it in production. The average computer user does not even use a computer as their primary computing device, they use a phone. The barrier to going from idea to phone app on iPhone or Android is relatively high.
Todo list apps, habit trackers, and the like are almost a special snowflake breed. Almost everyone has some different cross-section of needs they care about, and no app is perfect for each individual. So it's natural to say "is there something that matches what I want?" and then reach for tools to make that. The world is your oyster for todo list apps. Of course, the real issue comes from data sovereignty, trust, quality, things like that. When Apple launches a new device or a new iOS feature people want, you get to see which apps will actually implement the new features or which stagnate. They're a natural avenue for vibe coding since they're so particular.
It's somewhat like the Shortcuts system on steroids.
It would be interesting, particularly for Apple, as this would cannibalize fees charged on the App Store. I imagine they could charge for use of the vibe-coding platform, but Apple hasn't been great at figuring out LLMs.
It would be cool if 3rd partly app platform could provide this functionality, but as I noted in another comment, I cannot even install my own vibe-coded apps to my own iPhone. (Without the 100 USD a year developer tax.) So I'm not sure how the architecture would work on iOS.
They could have done a low-code/no-code platform ages ago, but they didn't. They don't want their users creating their own apps instead of buying, because that would cannibalise their income from hopeful app devs selling 5 units at a 30% cut.
People in general would recoil in horror if they knew how many essential operations are backed by a mess of Excel sheets with formulas and VBA nobody understands anymore.
All it needs is the maker mindset of being just lazy enough to be bothered by a repetitive task and the courage (and permission) to use an Agentic LLM to figure out a fix for the issue.
You end up with a system that works right up to the moment it doesn't and fails spectacularly and expensively.
This is one of those reasons you always hear about sweeping medical/hospital records systems being upgraded going tens or hundreds of millions over budget. The edge cases are demons.
Just normal non-coder jobs have massive amounts of repetitive crap that could easily be automated - and already has been automated with Excel - to a degree.
Now Agentic AI lets them automate the rest - or if they're real smart they use an agentic AI model and create an application to do it that doesn't require a LLM subscription.
Yeah, but that mess is deterministic! With a little bit of rigour, someone with no experience of that specific mess but knowledge of excel from a previous employer will dive in, make a small change, see if the results are messed up, back it out, try again with a different change, and repeat until they get what they want.
Good luck asking an LLM to modify something made by a different LLM 5 years ago.
Someone can accidentally type a static number in place of a formula and it may stick there for years.
Or Excel decides that a name of a gene is a date: https://www.theverge.com/2020/8/6/21355674/human-genes-renam...
Or Excel silently losing important health data: https://www.bbc.com/news/technology-54423988
None of these would've happened with a vibe-coded bespoke tool. A basic elementary level unit test would've caught them instantly.
Vibe-coding accelerates the destruction of basic (closed-source) apps charging a subscription for features that offer little to no value whatsoever.
Second of all, all of these SaaS apps that don’t actually have a need for recurring charge probably should be paid one time. I don’t use Loom — I use CleanShot X and it was a one-time $30 payment and has a lot of great features I benefit from. I can’t reimplement it in $30 of tokens or $30 of my time.
But for an app whose use case doesn’t change and is recurring for no reason? Yeah there’s probably not much value in recurring payments outside of wanting to support the developer. I pay a lot of indie devs out of the goodness of my heart, and I’ll continue to do that.
But the value for “SaaS apps” without clear monthly costs should have always been under scrutiny.
Perhaps LLM's will force developers/companies to change their stance and to stop users from recreating what they have already created, just buy an at-a-time snapshot of their app for a one-time-fee? Probably not but one can hope.
But everyone wants us to pay $10/mo. It just isn't sustainable from a consumer perspective.
And so few actually deliver $10/mo worth of value. If 1password and Fastmail - the two most important services that control my digital life - are each $60/year, that's the standard of value other SaaS companies have to beat and very few do. The ones that do are like NextDNS where they cost $20-30 per year because the people running them aren't greedy lemmings trying to pay back VC.
Of course it isn't. Just because some products or services are great value, doesn't make other products bad value. They can be anything from good value, to average value to low value.
And products / services are of course not comparable just because they are subscription based, or used on a digital device.
Gas has a fantastic value, one liter can transport me and my things a long way in short time. So does that mean that I can never buy a bottle of wine or some coffee outside of my home? They are after all liquids, and neither coffee nor wine can compare with the great value of gas.
Sorry, but no. If they're worse value than my email and password providers which my digital life revolves around and who only charge me $5/mo each, then yes those products are a bad value.
I pay $3,000/yr for Altium, $200/mo for Claude Max, $60+/mo for ad-free streaming, and begrudgingly $50/mo for Adobe so I'm not against paying thousands a year in nice fat profit margins if they provide actual value, like a shit ton of GPU compute time or a well made piece of professional software. "Value" here is obviously subjective relative to the beholder, but IMO the vast majority of SaaS I look at are hardly worth two bucks a month, let alone tens.
> The vast majority of SaaS I look at are hardly worth two bucks a month, let alone tens.
Then why are you looking at them,
I think your value system is completely broken if you think I can’t have a beer just because they cost more than fastmail. Some beers are better value than others but I enjoy having a beer. I don’t enjoy logging into some overpriced SaaS to do something that Claude can do for me now instead.
> Then why are you looking at them,
How can I evaluate their value if I don’t even look at them?
There are no subscription services which can beat Fastmail, iCloud, Kagi or YouTube in value for your dollar. So you can stop looking.
There are many subscription services which offer good or even great value, or mediocre. But since you demand that value has to be better or equal than the great value Fastmail gives you for you to be interested, then I'm telling you that you're not going to find it.
I don't order a coffee to my house because I can make coffee at home.
But if I'm at another city, then paying for a coffee beats not having it.
In that case, people who run Bitwarden for free are screwed. In fact, looking at how much I use the web browser Chrome, and how much I get out of that, and the fact that I pay $0 to Google to use it (inb4 I'm the product because I'm not paying for it), paying money for anything digital is terrible value!
What you've discovered is that prices are all made up. If we think about how to price a product, say a chair, from first principles, you'd take the cost of the raw materials, the time it takes you to turn those raw materials into the finished product, add a %age profit on top, and call it a day. In the real world though, that's not how pricing things works. You have a product, which costs $X in raw materials, and then you just... make up a number, $Y. Hopefully, $Y is much greater than $X, and you're able to make a great living off selling your chairs. Maybe you're called Eames and people will pay you $5,500 for your chair/lounger, maybe you're Office Depot and sell them for $129. Maybe you're not very good at chairs, so they're not level and then you can't give them away, not even to your friends.
Life is not an optimization problem. You can optimize for value, but then you'll find yourself in Walmart at 1am realizing that the 3-pack is cheaper per-roll than the 30-pack that night for some reason, and getting angry over that.
That's not at all how I think products are priced. That sound's like something you'd tell a kindergartner to shut them up.
> You can optimize for value, but then you'll find yourself in Walmart at 1am realizing that the 3-pack is cheaper per-roll than the 30-pack that night for some reason, and getting angry over that.
I have never found myself in a Walmart at 1am* nor have I ever gotten angry at toilet paper (I get the Charmin ultra from Costco like a normal person). You need to re-calibrate because you sound like an Inland Empire methhead. Pro tip: you want to shoplift the detergent. That tends to trade better with the other methheads.
* Not entirely true, but that's just because the Reno Walmart stocks up on Burning Man supplies and Gerlach only sells shitty playa bikes.
I assume you use vintage fire hoses from the Birmingham campaign to clean down there?
It's just common-sense though - if the market is willing to pay me $100/widget, why would I sell it for `($10 cost to manufacture + 35% markup)`?
The new lower bound for simple side-hustle apps now is virtually zero. All you need is a computer, electricity, internet and (optionally) $20.
Agreed, but the confusing part is that you don't seem to be saying "to me, those services only provide X amount of value, and I'd rather have $Y than that" -- you seem to be saying that if 1password and Fastmail were more expensive, you might be willing to pay the asking price for some of those services you currently consider bad value.
But they don’t, which is why I use them as my baseline. If my email provider and password manager - two services with damn near infinite vendor lockin - can do it, no one has any excuse.
>I pay $3,000/yr for Altium,
You are definitely a professional/industryspecialist - so its quite obvious why you pay for this, same as a mechanic is paying for very important and expensive tools.
This mindest is not possible for "non-professionals" :-)
Funny you mention Fastmail. I was happy most of the past decade until this week. I just had my email blown up by their new Paddle billing system with a ton of billing invoices since they decided it was no longer ok that I pay them a lump sump every 2 years, and that I must go onto monthly now. Initially I thought they were hacked but nope, just terrible communication.
I emailed them a few days ago and they only confirmed that Paddle is their merchant of record and they have been migrating accounts over slowly.
Tonight the CEO sent out a blast saying resellers need to be on monthly billing with their new system at new pricing.
Sorry Fastmail, I paid for 2 years back in October (I think this is my 3rd cycle with them). If you want me on monthly billing then you will wait until October 2027. That is a ‘you’ problem not a ‘me’ problem if you undersold the subscription this cycle.
The BIG difference is that we didn't felt entitled to use everything that is fashionable or switch apps every couple of months.
We would do a research across several magazines, local computer clubs, and the few lucky ones that had online access, some BBS or Usenet groups, then buy that one package and live with it for a couple of years, regardless of their limitations.
Instead, the reverse has happened and platform churn has risen to new highs, necessitating subscriptions.
Eg in desktop OS's. Apple for example makes everyone miserable by re-breaking macOS every year. To what point?
... Even for desktop Linux users? I can't say I've felt it. I switched almost 4 years ago and it just keeps feeling better and better (in a "Luigi wins by doing nothing" kind of way).
Would be nice to see the XDG stuff like portals etc. better respected, though, yeah.
The only reason why Linux desktops work at all is thanks to package managers and their maintainers doing the heavy lifting of keeping applications and the libraries they use in lockstep. If it weren’t for that random programs would be breaking every other update.
They are, but it's not their fault - Wayland removes so much functionality that X had, and delegates that to the WM/DE.
I tried to do a small personal app last year for myself that intercepted and injected events for keyboard and mouse. Not possible in Wayland (so I switched to X instead) - that is delegated to the WM/DE.
There's hundreds of these tiny little cuts that cause friction for app devs.
Part will be for new features, but no doubt that another big part will be for platform support over time. There's just too little backwards compatibility guarantees nowadays from the big players. We need more Microsofts in that sense!
The code is the easy part but there's ongoing humans needed to make it work. If Agents get to the point they can genuinely autonomously SRE & patch a service everything changes but that still seems a long way off.
But that might require just firing some people because that amount of man-hours is not needed any more or moving them to make something new and no investor likes it
For $40 product, I will rather pay $1/month than $40 once. It keeps the incentive aligned. I think most people just assume that if the devs move away from subscription they would be fine with lower revenue and would charge less.
Does the software stop working after the 2 years? If so, I’d go subscription (or find another product that doesn’t explode). If it doesn’t, I’d pay the $400 assuming I want to use it for more than 2 years.
> So let's say average subscription time
It could be anything from month to decades depending on the subscription. And the price should be accordingly (risk based)adjusted for the same revenue.
How would the economics of this work universally? Jetbrains is a bit of an oddity in terms of SaaS. For the most part, it's desktop or on-premises software that was sold with a perpetual license. If you've bought a subscription and canceled, they've generated some revenue. Maintaining a subscription generates more revenue for them, but they can slow or stop development without stopping you from using the product.
SaaS is typically some server software hosted by someone else which most often doesn't have an on-premises version. They can stop making feature updates, but if they turn off the servers, the service ends. They still have costs even if you use the software less. You can argue about the profit margins, but that's not the point here. As most SaaS companies don't start with on-premises, they can't ever get their software working there for many reasons. There are a few like Atlassian and GitHub that do both, but if you look at the heritage, both are really on-premises first.
I vibecode an app that only I use and store data locally. That means my data never leaves my device, I never have to share my email with anyone, never have to enter my credit card info anywhere
You buy SaaS and you have to then login, share credit card info, and have your data stored in the cloud somewhere with godknows what security practices
That’s worth more than the cost of any tokens
So much infrastructure is there to support doing "it" in the Cloud, for all definitions of "it." If we can vibe-code bespoke one-offs to solve our problems, a lot of that Cloud interaction goes away... And that stuff is expensive and complicated.
Hypothetically, open source app stores (I'm counting apt here) address this, but then it's someone else's solution to my problem, which doesn't quite fit my problem perfectly.
This approach to software engineering could be what 3D printing is to tangible artifacts (and I mean that including the limits of 3D printing regarding tangible artifacts, but even still.)
I'm speaking more about fake SaaS "we have an app and we charge monthly for it and license it." Obviously, a tool with cloud-based storage and sharing will be a different beast.
But do you trust a vibe coded app to do cloud-based storage and sharing better than a company or an indie developer? If you need these functions (like sharing a todo list between two users), you have a lot more concerns than "does it boot".
But that's cheating because I do this stuff professionally. Would I trust a vibe coded Todo app my uncle's son Jimmy who smoke a lot of pot uses? I'm not saving my bank account number to it, but I'd have no problem using it for reminders that my aunt's birthday is coming up so I need to buy a gift. If it gets popular within in the family unit, uncle will have me talk to his son and take a deep look at it anyway, try and encourage Jimmy to go back to school and look for a job and all that stuff too.
Plus, theres nothing stopping the company and indie developer from vibecoding as well.
I mean, if you vibecoded it you don't actually know that, do you?
Can’t do that with SaaS
Also, I’m baffled that on HN of all places, I have to actually defend the idea of rolling your own apps and protecting your data from cloud providers
If only
Especially with agents that slurp up files, have access to databases, etc. You're literally giving access to your computer, your network and your data to third parties and letting them run code.
(If there is a bug though which I would want to debug and if I were not a developer, then your concerns are more valid)
I’ve seen projects where testing is done in prod and also projects where API keys for some external services (e.g. Mapbox) are shared across prod and dev. Or cases where credentials end up in Git repos (edit: wrote GitHub originally, but meant typically non-public repos on any platform) due to ease of use and how inadequate secret management solutions can be.
Luckily that’s not the majority of projects, but I bet it happens a lot more elsewhere. Definitely a bunch for your average outsourced/freelance/scrappy/non-funded project.
Whether anyone will actually use your secrets or even code that’s sent to these large AI shops, though, that’s another question. You might as well question using GitHub cause it’s owned by M$.
What is valuable is the data you store in these things. Ergo, it makes more sense to vibecode these because the code is routine, but the data you save in them is worth protecting
It should have been, but the number of people qualified to offer proper scrutiny has been low, and those people have largely been occupied with bigger things. Or they were making those apps and had a conflict of interest.
The point is that now that vibe coding is at a level where it can identify and put together off-the-shelf components, and there are all these end users that don't really care about standardization (it's not like their SaaS products used open, interoperable standards in the first place), the ability to compete with those offerings has exploded.
if you want future updates, you pay for another year of updates (discounted, of course, for loyalty).
or is it more compelling to just have one clean, flat, lifetime rate?
The industry had arguably more innovative products than exist today and that business model worked totally fine until the platform gatekeepers and VCs invented SaaS because they decided they weren't making enough money and needed to do some rent seeking.
unfortunately, we no longer live in an economic environment that supports being able to run a stable business that way. if cost of living were still relatively stable/affordable/predictable? sure.
i would love to see everyone go back to this model, but sadly i don't see that happening unless the person offering it is independently wealthy/comfortable by other means.
A few recent example purchases (macOS): BBEdit, Base, A Better Finder Renamer, Nitro (that’s lifetime, though I actually prefer “until the next major version”).
The developer then creates version n+1. The old version is kept supported, but new features go only into the new version, which you can optionally buy again.
I wouldn't charge customers _less_ for that just because it's now a one-time payment.
You can pay to unlock advanced features and keep them and any new features added in a year, after that any new features are paywalled for another unlock, and another +12 months, perpetually.
At least for games I think it's much too late for the one time purchase model unless you're part of a pro studio making relatively big games.
and when whatever framework or big lib you use decide "well we're making new version, everything you made will break, have fun", that needs engineering too.
Buy once model is pretty much only for standalone, offline apps. Anything online and you have to start to worry about supporting new TLS versions or having to update certificate store (if app for some bizzare reason ignores system one)
You pay less tax to them when you do subscriptions rather than one time payments, if I recall especially with Apple Store (something like 30% first year, 15% the second year of subscription).
This is why 120$ over 10 years (1$/month) is more profitable to companies than 120$ in a shot.
Yes, but it's almost one time payment. Your own personal use case is usually narrow enough, and you don't need to support different OS/browsers. You can vibe code it and just forget the fact it's coded.
Actually it's the best use case for vibe coding (the strict meaning of this word) - when you don't plan to maintain the codebase anyway.
Yiu don't pay taxes on the time spent on your own projects.
There is no alternate I've cost, as the time would usually have been spent watching television if not doing these projects.
If you are dedicating a considerable amount of time on a project, there is of cause opportunity cost, since that's time
- not spend on improving health - not spend on improving skills - not spend on improving other projects
Seldom one decides between a side project and starring into the void
Probably not.
I also paid for FreeFileSync (the donation version) instead of rolling my own local backups (across drives), as well as MobaXTerm because I really like their UI/UX and so on. Software that others have developed and supported for multiple years, which has been already tested by a lot of folks thoroughly across their usecases is probably a good bet, doubly so if you can buy it (or I guess choose to support the devs) instead of renting it, the difference being that in the latter case you're not in control.
At the same time, 30 USD gets you about (assuming 85 input and 15 output split, which approximately matches my stats across months):
* Gemini 3 Pro: 8.57M tokens (7.29M input, 1.29M output)
* GPT 5.2: 8.36M tokens (7.11M input, 1.25M output)
* Sonnet 4.5: 6.25M tokens (5.31M input, 0.94M output)
(actual figures would change with the input/output proportion, it matters a lot, and also any caching)
Not really enough to build serious software, but definitely quite the bit of help along the way!
Or if you go with subscriptions, it can vary even more, even if will cap how much you can do per day.
For example, if you pay 50 USD for Cerebras Code, you get 24M tokens per day, so that'd be close to 730M tokens per month. They're running GLM 4.7 which isn't SOTA in my experience, but is somewhere around Sonnet 4 and therefore actually quite capable: https://z.ai/blog/glm-4.7 (my experience might not match the benchmarks, but either way it can be good enough for most stuff out there)
For a decent percentage of software development, maybe where you want a feature nobody else out there has, AI can help you get rid of enough friction and lend enough help along the way, to maybe make it worthwhile. The caveat there might be that you have to treat the expense of your own time as something you do for the enjoyment of it (building something, or the delayed gratification of benefitting from using the software).
And it is getting better at a pretty rapid pace.
If you need $1M/y in subscriptions to build your software, you'll be outcompete by solos who needs $60k/y and don't care about the 100% churn of a one-time fee.
This is simply market optimization when the marginal cost of producing the good falls.
Some of these things are the feeling of "getting a deal" or "screwing the system" and people can spend their time in whatever way brings them joy. I don't want to be a "markdown editor app developer" or whatever and would rather work on more creative endeavors. In the end, we give our time to get money and then trade our money for time but time is our only limited resource.
More generally, I don't view time as some resource to be used to achieve some ultimate goal. In the end the only goal we're going to achieve is death. So in the interim I'd rather do actions that give me satisfaction rather than those that give me money, because in the end - the point of that money is to bring satisfaction anyhow, so I'm just cutting out the middle man.
Sure, if you hate coding it's prob not worth it. But I don't see it as an expense. In fact I value it so I would even not go as far as to deem a failure as a loss.
Previously, if you wanted to rig up a screen recording app in a few minutes you'd use Linux and specifically one of the minimal desktop environments like i3 or sway. Then you'd wire up a few command line utilities (slurp, wf-recorder, etc) and get the exact experience you want. These small unix tools allowed the greater ecosystem to just cobble together what they wanted without too much hassle. I've done exactly this!
This is the exact mindset that keeps people away. They'd prefer to NOT maintain their own video recording software THANK YOU VERY MUCH. So you get people who go to Mac OS because there's always a slick solution for a couple bucks and you never have to be TOO fiddly.
This doesn't destroy the market for people in the latter camp, but it DOES open up the market for people in the former.
And this is comparing to being subscribed many years in a row. With SaaS you can unsub and sub only when you need it again.
With your side project - a weekend of your life is invested and you will never get it back.
This is the worst use of your time if you measure it in $. If you make it for fun - sure. In all other terms it is a complete loss.
Hardly doubt that this was the 'most waste of ones time'. For one, it's not like most of us can decide to "work" for 3-5 hours on a Saturday and get any money. I play games on my pc while claude codes for me. I alt tab each few minutes and see if it needs any input. Then I can (not that I do it), read and perhaps learn from the code.
(How? Idk, I just asked it to guide me through the short hook process)
I get a notification on macOS with the title of the context.
Subscription apps often have to target a wide userbase. However, most users only need a small subset of the entire feature set, and would be better served by a tailored version. This means that vibecoded apps can get away with being much less complex (specific featureset, no login etc), while still being more useful.
I have also created tools with LLMs that are exactly tailored to what I need, and still much more polished than what I could do without LLMs. Will have to think about if there is anything else I can do this with.
Once again, it took me about an hour while watching my shows to get a custom one made.
The first version operated by me downloading the pages one by one to a directory, the Python app parsed the html, downloaded the files and renamed according to thread name.
After a few iterations the tool just grabs a cookies.txt file exported from Firefox and can take any thread URL, browse through it, skipping existing files and determining if everything is already downloaded
I could easily have it just watch a set of threads for new content and download automatically, but the current system is fine =)
I need a simple S3 compatible API to store some files with basic auth and ssl certs using let’s encrypt. Nothing crazy, Garage is overkill, Minio is overkill. I may see if Claude code can handle that for me using python or something.
/btw, I work in consulting and the above project would have a budget of probably $100k and a schedule of 3 months. I see a lot of change for swe consultants coming.
If the commercial provider charging you $10 a month breaks it, you also have no capacity to fix it.
Your options are: send them an email, or unsubscribe and use something else.
But if they remove a feature I rely on, I can't put it back.
If they add a feature I hate, I can't remove it.
If they jack the price up, I have no real solution to this.
If they move features I rely on from the standard tier to the 5x more expensive pro tier, I have no real solution to this.
Why, yes, this is an echo of the old argument for open source software.
Now crap has turned into revenue sucking subscriptions, at most yearly licensing, feature flutter. And the worst is being bought up by VC/PE and milked for anything useful and thrown away.
So much this. Each subscription is literally a small percentage of your revenue. You can't reinvest it ... it's just gone. Hopefully it enables more productivity ... but most likely, it is only marginal.
In what situation would a simple script or helper app just suddenly rot away and stop working?
Of course it's POSSIBLE to vibe together a massive monstrosity of an everything-app, but that's not what the author is doing here (nor me).
With a agentic llms I can just tell it to fix it. With a commercial solution I'm fucked and either have to find something else or pay for a license (or keep paying every month).
cheers
From the repo here, it looks like its just using swift command line tools, which might just work well enough with cursor/vscode/etc. for small projects. You won't have Xcode's other features but maybe thats fine for an agentic-first development workflow.
We can now produce products and apps that are tailored to our own preferred ways of working.
Regardless of the cost of generating them (which can be as low as $20 per month for a ChatGPT Plus subscription) or the effort involved (sometimes less than an hour of “vibe coding”), we’ve reached a point where the resulting product can be significantly more valuable than the existing product, service, or subscription it replaces.
That's way too much money. The opencode default models are free
That’s the whole point - there’s no need for it to be a product when you can do it yourself, and it’s the death knell of products like this.
You can just use it and be content.
FWIW, `better-auth` solves that and LLMs know how to integrate it.
At the same time, Apple better wake up as the quality of the App Store (such as it is) is about to get much, much worse as a flood of vibe-coded apps make it onto the store. These will likely crap out in spectacular ways and, as I say, bring user's experience with the App Store in general way down.
It's like when Amazon intermingled drop-ship sellers on their platform.
(Oh well, not my problem.)
List of projects mentioned before: https://news.ycombinator.com/item?id=46716805
But Roberto's use-case is definitely more sane than most.
However, with a proper framework (e.g., a very opinionated design system, the ability to choose from some pre-designed structures/flows, etc.) I could very much see ad hoc creation of software becoming more widespread.
Their problem is solved, now it's up to me to update the internal guidelines and agent instructions so that the code is at least semi-decent.
None of these are going to "production", they all live on local company controlled laptops and only one of them might access an external API automatically later this spring.
But each of them takes hours of manual work and does it in minutes.
But typora is actually one time purchase and one of the rare apps that is priced well with good business model.
They have probably best RTL support and I wanted like your friend to write my own focused markdown editor with RTL support using clause and made some progress but realized that the time and cost of doing this is not worth it. I just paid typora a week ago for $15.
But I understand the point and I use Claude to hack together personal tools all the time.
Now I wonder if the maintenance cost for this type of internal system has gone down to a level where that is no longer an issue.
Or a vibe-coded simple website.
But "designed and implemented company-wide intranet" looks good in someone's CV so here we are.
Here's what bugs me: I cannot permanently install my apps to my iPhone because of Apple's walled garden. I need to reinstall every 7 days and constantly re-confirm that I am a "Trusted" developer.
I know I can pay Apple 100 USD a year for a developer account, but I bought this phone outright 7 years ago, I own it. (Obviously, I clearly don't in this case.) /rant
I wanted something that felt like an app, so would use iOS design elements, have widgets, use on-device storage (for offline use), etc. Apple, very intentionally I believe, makes a lot of these things harder than they need to be.
Does that work better for maintainability than letting it decide on its own what the architecture should look like?
If so, what is your setup/workflow?
The cynical approach would be to make the product hard to leave. But that just means you've built a trap, not something people actually want. Eventually they escape and hate you for it.
The test I use: would people recommend this to colleagues even if there's no referral incentive? If the answer is no, I'm probably building something people tolerate rather than something they value.
Where I could see some really interesting results is the marriage of the two. For example, you have a solid data structure that an LLM can generate infinite custom views from.
And I have first hand knowledge of well-known companies building their own tooling because the SaaS offerings have a bad price/feature ratio.
but from good apis, good data, good interface they can generate quite nice frontends.
i guess, frontend as job is going to have a hard time.
also, writing code is not cognitive load, its always reading code. and llms just increase that. so i mostly try to avoid using them.
but i do like researching with them. context free. like googles ai mode, etc. not from my code editor cause then they get biased and suggest stupid sh8t all the time.
The recommendation thing is a nice benchmark, but if you're building hyper-specific tools - why would people recommend them to anyone? If you build a tool for an accountant that does some very niche thing only they're bothered by, why would they recommend to the analyst or receptionist in the company?
I replaced a whole bunch of these with one shot prompts for shits and giggles.
I vibe coded a webapp that I was paying yearly for and the version I made does everything I wish the app I paid for did as it's 100% personalised to me.
I've been thinking for awhile that this is going to be the future and I'm already starting to think of more things I will create.
I haven't published it publicly yet, as i use it personally and it's a little flakey still, but will look to do in the future once I finish adding all the features.
My version is much more feature rich than TeuxDeux and I made it for Free in Google AI studio over the past two months, between other things. I'd just type a prompt and then go do some other stuff, it has taken quite a lot of revision but I haven't written a single line of code and i've been using it daily to manage my tasks since the start of December.
Building all the software you use yourself, whether by hand or by vibe coding, cuts you off from the world.
I have no philosophical objection to vibe-coding apps for yourself, but personally, I wouldn’t be 1/10th of the engineer I am if I wasn’t constantly exposed to the work of others.
For some, this trend worries software engineers — who needs software if they can vibe code it themselves? — but I am much more optimistic. I think people will start valuing good software a lot more. Claude code can deliver the first 90%, but we all know it is the last 90% that differentiates.
I switched from Spotify to buying MP3s and using my own audio client, because I'm fed up of a company telling me which music I should listen to every single time I open the app. It costs more, but I own the music and I escape the constant redesigns, price increases and influential behaviour.
Most apps are very simple and there isn't too much to learn, especially if you're building it to scale to a userbase of yourself. I can't see the need for a ton of CRUD apps which demand subscription fees personally. If you build them yourself, you get to keep your own data, build it out the way you want it, keep it that way, and use computers as a person using a tool as opposed to a customer buying a product.
A minimal web client audio player with some basic database tables in the back for organising and searching does me fine.
Decisions matter, both technical and product ones. LLMs don't make as good technical or product decisions as I would, and the way I work with them tries to maximize my strengths and the LLM's strengths. I don't know if I succeed, but it's better than "make me an app like X" as a prompt.
SWE field is one of the most cognitive dissonant social groups; cries foul at the slightest whiff their free speech and agency is being put upon; seeks to reduce blockers to their productivity, fewer PMs! Less management!
Now complains about users using their machines without having to block on an SWE.
Insert that quote about how someone will not see the obvious if their paycheck relies on them ignoring the obvious.
Here come LLMs and all they can accomplish with a few arithmetical rules instead of the arbitrary semantics of an SWE; watch as SWEs block social evolution away from disrupting software engineers.
As an example; "protected memory", among many other individual software problems, is an access control problem mired in old semantics relative to OS monoliths.
Didn’t see you all halting as you decimated travel agent jobs, retail jobs, etc etc. Technology advancement must now stand still after centuries of evolution? The self selection bias is as obvious as Trump's.
You have to remember that SWE's are the same group that screams "communism" the first moment you mention the word union and they should have the right to make as much money as possible with no restrictions.
This of course leads to the obvious lack of self reflection in their responses when something threatens their future income.
I'm not a SWE because I like money. I'm an SWE because I love programming.
these tools have been trained on decades of people "obsessing over every last detail". what GP is arguing is that we're detaching from that: you prompt, you get something that works, it doesn't matter how it got there. we're now entering the world where the majority of code will be vibed. So whatever our foredevelopers came up with, that will be the the final chapter of craftsman-produced, understood, code. whatever the previous generation actually learned about software engineering, that's at an end too, because why bother learning when i can prompt.
there's no stopping this transition, obviously. the next generation of tools will be trained on the current generation of tools' generated code. we're passed the "termination shock" of sofwtare understanding.
Am an EE and have argued against all the developer gibberish and self aggrandizement for years. It's just electromagnetic geometry of the machine to me.
Most software out there is all the gibberish devs need to do their job. Burns a lot of resources clinging to it. Completely useless to using a computer how most users will.
Vectors as a uniform layer of abstraction, rather than arbitrary namespaces a programmer finds cheeky, will obsolete a bunch of gibberish.
That said, to address your broader point: me, personally, I am thrilled that the barrier to entry for building software has been all but eliminated. The joy of creation belongs to us all.
Building all your transportation yourself—whether by breeding horses or assembling a Model T—cuts you off from that accumulated experience. You lose the benefits of thousands of hours spent by others thinking carefully about the same problems.
I have no objection to Model Ts for personal use, but I wouldn’t be one-tenth the traveler I am without constant exposure to well-bred horses.
Some worry cars make horses obsolete—who needs breeders if anyone can buy an engine? I’m more optimistic. As cars proliferate, people will value good horses more. A Model T gets you the first 90%; it’s the last 90%—judgment, robustness, and adaptability—that differentiates."
No one is doing that. In foreseeable future I don't see people making their own OSs, browsers and drivers. Workplaces never ditched Offices and Windows for the open source counterparts and they are certainly not going to do that for vibe coded solutions.
You can rest assured.
How will people determine what is good software and what is not? Even experienced engineers can't tell just by looking at the final product.
Some of the most solid rock-solid applications I see were built years ago and still look primitive (native Windows 7 controls, etc). Many of the worst, bug-infested anti-user software looks slick and modern.
> Claude code can deliver the first 90%, but we all know it is the last 90% that differentiates.
My experience with trying to complete that last 10% of a CC generated project is that it's all very alien looking; very uncanny-valley vibes, and I have serious velocity issues because of the lack of coherence.
So most software is 180% of 100%? :p
Not everything has to be a SaaS, but I don't think all SaaS apps can be vibe-coded to a weekend project.
If it is solving my issues and problems, why do preaching about the merits of a proper product or paying. I'll pay for what I see value in, and vibe-code where I don't see the benefit of paying.
Maybe I miserably fail and get back to paying to product. It's all good, I take that responsibility while I start my vibe-coding session.
Once you identify something like Handy, instruct Claude to study how that OSS project actually builds the feature and adapt the logic to your stack. AI is really good at finding the "seams" (those connection points where a feature ties into the tech stack) and understanding the full implementation.
The trick is knowing precisely where the feature lives in the code (files, functions, modules), because AIs often miss scattered pieces and don't capture everything otherwise. That's what I'm working on at opensource.builders[1]: turning OSS repos into a modular cookbook of features you can remix across stacks, with structured "skills" that point to the exact details so the porting works reliably.
My projects (Openship/Openfront[0]) are the first on the chopping block. We're creating modular OSS alts for every vertical (barbershops, hotels, etc.) for folks to take, remix, adapt, or fork into their tools. Chances are your AI model is already trained on similar OSS and building from it anyway. We make finding the exact code reliable. Check out our ethos to learn more [1].
We have so many people who are so excellent and fast and developing nowadays that we can even afford the time to build things for our community, friends and even just for ourselves.
It has probably always been like this, but I am just personally observing a higher-degree of people doing and talking about it. Even just the small-web/neocities bobble points into this.
Running out of "Pro" tokens just means, "Time to set the coding aside for today and get a few other things done around the house."
Now that Prometheus (the myth, not the o11y tool) has dropped these LLMs on us, I've been using this thought experiment to consider the multi-layered implications:
In a world where everyone can cook, why would anybody buy prepared food?
I would guess for convenience and saving time. While vibe-coding might be faster, you still have to "do it". As in, think about what you want your software to do, write out or dictate your prompts, test that it works etc. That takes time (might be less time than writing it out by hand but it's still a non-zero amount of time).
Because of course we all buy prepared foods of all sorts from street vendors to fast food to local restaurants to chains to Michelin Stars. While there are many reasons one will cook for themselves, there are many reasons one will buy from someone else too.
So I decided to vibecode an app for myself and wouldn't you know? It took me a few hours and it's INCREDIBLE! No more relying on someone else to maintain something, I can simply build my own solutions, whenever I want!
Other thing I have experienced is my standards have changed a lot, now for $10 subscription I need a lot more, not just some simple editor or a small todolist would suffice anymore. I am not thinking about paying for new software, and in fact I am getting completely burnt out by all the sites looking the same.
End of the day, much like when photography went digital (and smartphones got good cameras), yes, there were a LOT more photos taken, but the relative proportion of outsized, lauded photographers remained fairly constant. The upshot is that WAY more people are exposed to the possibility of creating excellence than before, the downside is the market gets flooded with utility and mediocrity. Said excellence never goes away, and the same will apply to software.
The very idea that SaaS (or packaged software, or whatever) "will die" because "anybody" can prompt their way to a "personal tool" (as a mainstream exercise) is so far-fetched to me because the only people who will prompt their way to a tool ARE SOFTWARE DEVELOPERS!
Professionals who need functionality will always pay for it.
Boomer dads who can barely work a DVR will always pay for it.
Business owners who need less friction and more reliability will always pay for it.
IMO, this "I'll just replace Salesforce with my own personal CRM for $200 for a month of Claude" thing is just a hobbyist's pipe dream lol -- not that there's anything wrong with it, some people will do it, but, man, there's a reason that Netflix is Netflix, and Plex isn't Netflix.
I don’t understand why Netflix needs 3500 engineers. They built what needed to be built already.
I feel like if e.g. Hypercard had lived, this would be a more defacto mode of doing things.
It just boggles he mind how divorced from reality some people are. You could offer $fotm_ai_model with infinite usage, free apple developer account (since you're apparently replacing everything you have with homegrown stuff) and the amount of people wasting their weekends on vibe-cloning their own custom apps would still approximate to zero. This doesn't even get to the fact that the majority of apps already HAVE a free alternative, and it's certainly far less effort to replace increasingly obnoxious apple music with foobar than to build, test and then permanently support your own music player. You also probably want claude code 15 to replace anything non-trivial, otherwise, well, good luck.
I don’t think I’d bother even if my weekend had ten times as many hours as it does, and I’m a code monkey that still mostly enjoys his job.
Linux is free, but most people don't mind to pay the Windows / macOS tax.
He turned an ongoing expense into a one time expense. And bonus he likely had fun doing the labor.
(1) Let's hope he doesn't cut himself if the knife breaks apart while in use.
(2) Why is it so hard to just buy knives these days? Why does every knife maker suddenly feel entitled to charge rent?
I'm all for DIY instead of hiring others, but the insanity here is the rent vs. buy thing going on in the industry.
And calling Windows and MacOS a tax is a complete misunderstanding of what a tax actually is.
People like to repeat this thought terminating cliché because they think it makes them sound smart and insightful, to doubt that a free thing is really free. But it's an uniquely naive opinion.
On Windows or MacOS, it's more often than not that, when you meet a problem, the only thing you can do is throw your hands into the air, and suffer the problem. This is paying with your time. Every single time you have to sit and wait through a forced update, this is paying with your time, in the realest sense of the word. You give your time for continued use of the product, and nothing else.
What people mean with the idiotic folk-wisdom is that you spend lots of time with the internals of Linux.
That's not true, but let's assume it is. The internals of Linux are likely a thing you'll really want to learn in depth if you're professionally into any computers science related job, because the market has settled on that.
If you're not, it's also something you'll want to learn, because Linux's design makes a single skill you learnt applicable in a lot of different workflows. So stuff you learnt while troubleshooting, you may adapt in other situations.
You don't pay with your time, you invest your time. Like all investments, it has an initial cost and dividends. It's a pretty good investment.
You don't fiddle more in Linux than you do in MacOs or Windows nowadays and that screen recorder tool he had to spend time vibecoding is embedded in the desktop already, the rest of the apps are 2 clicks away from being installed.
Game over, devs
I think it's more likely the case that AI is not going to be as disruptive as they think.
I can hire a contractor to build a carport, or whack one together with some supplies from the Big Box store. More roofs being built with more price points to serve the market.
My wife was preparing for a specific exam and found an AI-powered mock exam app. The trial version was polished, but after she paid for full access she realized most contents are of low quality and some basic functionalities are broken. I imagine with LLM one might easily create such lemon products in infinite niche markets even without much domain knowledge, scam out some money from new users, and care not about survival.
I wonder why people still hold a lot of stigma against something that was built assisted by an LLM.
I am rewriting my website. I was using a converted Pelican template. I started the rewrite using variables similar to the template, then about halfway through, I realized, "this is dumb. I am the only user. I care about nobody else. I can hardcode nearly all of this, and if I want a change, change the hardcoded name." An example of this was various social media names.
I have scripts that convert color themes for applications from more popular themes to a theme I particularly like. I hard-coded the input colors and output colors. I could have made a config file, etc, but, that adds complexity and, more importantly *I do not care about other users.*
There is a huge leap between "good enough for me to use for exactly my use case" and release or sell as a product.
- A nice little single-file web "random slideshow" to replace an aging one I bought. - A fairly feature-complete read-only SQL console. - A development SMTP server (like Mailhog) https://github.com/linsomniac/smtphotel - A work status dashboard that I'll probably release once I have run it a bit longer. - A fairly extensive Docusign-like webapp. - A retrospective meeting runner. - A cron "swiss army knife" helper. - A "social calorie tracker" (I'm unhappy with the existing ones out there).
These are all things I've vibecoded in the last month, and are more than I could have coded in my spare time in 6 months or more.
For me, the renaissance is here.
The other thing is, LLMs tend to generate terrible code that pisses off open-source maintainers. So I'm not sure even LLM-made patches will make it into open source much.
This might be the death of traditional open source. Vibe-coded-only open source may be the next generation. Which I'm fine with, as long as we can start regulating software, so that vibe-coded tools are banned for safety/privacy uses unless they follow a software building code.
If you aren't working 24/7 while handling a family and telling yourself your time is worth more than a small fee, you are just being lazy. I'm the same way, I am incredibly lazy and will constantly tell myself that my time is worth more. This is usually until I realize I'm spending way to much of my "money" to "save time". HourlyWage(time) = money, if I'm saving time by spending money I'm losing time. This is a basic concept and I defy anyone to show me otherwise.
We live in a time where instant gratification is the main driver behind most decisions, devaluing our currency each and every fee we succumb to... as money is time, and if time is being "saved" by spending time (in the form of money) we are now applying a future debt to the work we are doing today. You might work 40 hours one week, where at least 4 hours of that week goes to paying your streaming bill, another 8 for Internet and Phone, as well as another 2 for the coffee you didn't make that week, another hour for your notetaking app on your phone, 30 minutes for your subscription to watch funny youtuber release content early, another 2 hours for you glut of productivity apps, etc. These things all work to keep you a wage-slave till the day you eventually croak with a menial 401k.
It's embarrassing we reduce ourselves to this.
If people enjoy spending their unpaid hours building clones of paid software that's fine, but it's fine because they enjoy it. It's not minimally worth it. The time I waste on YouTube and the news etc etc is sorely needed and enjoyed downtime. If someone has enough energy to build instead of vegetating, more power to them. I prefer to save my energy for the stuff I value. (Which is actual work, helping family and games)
EDIT: another thing to consider is that each hour I spend fully pursuing my occupation pays me an hourly wage but also pays me in career growth. This compounds massively over time in higher and higher wages. Building throwaway apps generally does not. Why would I waste energy on work that doesn't compound? I'm all for serendipity but not as a financial argument.
Most people’s free time is not worth anything monetary. Sure, there is non monetary value in chilling out and spending time with family, but if I spend a few hours making an app for myself, then I got something for free.
Today on /r/macapps/ there’s a wave of apps that look good at the first glance but get abandoned before they achieve even a 100-users maturity level.
You know how you think and know your goals so a customized, focused solution would be better than something for the general masses.
Did I hit a nerve or do you just have a crush on me? :)
What made the difference? For one thing, I know better than to ignore the future. A small success in 2020 was like a seed planted amidst an infinite future, and I knew to water the seed. Secondly, I continue prioritize this fulcrum between complexity and utility, where the app gets better by getting simpler wherever possible (most people neglect this fulcrum as not offering much “business opportunity,” failing to realize it is the foundation of all business opportunities). Finally, I just love it, and appreciate that others love it too.
(For those interested, you’re welcome to join the beta at minimal.app/#beta on Apple devices and contribute to the roadmap.)
And then put it on an app store and put all the vital features behind $15/mo subscription.
Which is totally justified! I understand the time and energy needed to get a product polished for 10,000 users. But thanks, I will take my vibe coded one.
The bar for me to pay for a $10/month software subscription is pretty high, but once I make the decision that it's valuable, the actual cash cost is pretty low. Vibe coding something will never approach the quality of something that someone put enough thought and effort to turn into a product. The main place where I'll write my own software is when it's truly custom to my own needs, AI is a force multiplier for this type of work, but at the end of the day I still have very limited time to run and maintain a lot of custom software and data, so it's not going to cannibalize any of the SaaS I'm willing to pay for.
Obviously for younger software guys with more time than money, the equation will be different, but those were never the make-or-break demographic for SaaS anyway. I don't think the equation has meaningfully changed for SaaS sales due to AI, I see it more as continuously rising bar over the last two decades due to UX expectations, market saturation, limits on human attention and complexity tolerance.
I build small web applications for my personal needs all the time by just regular programming, and I'm saving so much money by using them and not some proprietary app. Not even mentioning the advantages that it is completely bespoke, runs local and gives me peace of mind data-wise.
Some wise man once said that personal computers are a bicycle for the mind. Programming your own programs is the most pure way you train on that bicycle.
That's sad for you. Do you spend the other two weekends dead for tax purposes?
Might have a bit of difficulty naming it that (https://en.wikipedia.org/wiki/Jabber.org).
Edit: I apparently wasn't at all the first to think of this (https://github.com/rselbach/jabber/issues/5).
But that’s not what Loom is about.
It’s about streaming the video.
Before:
Capture something with likes of quicktime.
Transcode it so that it doesn’t take a few gigabytes. This takes considerable time and resources (though OBS can do it while recording, not after).
Upload somewhere to share. Wait while it is uploading.
Loom takes care of all those steps so when you press stop you can immediately share the link with someone.
—
Hope other use-cases in the article are not as misrepresented as this one
But then they did made a bad choice for paying for loom. They could have just learned (or used llm) to make a bash script to use ffmpeg for capturing screen to a file. Or OBS is a pretty good solution as well. And a ton others
It's always been free, but because of a change to the way Amazon charges third party app devs, they were going to start charging next month. Since the whole app is just a couple of API calls and storing a record of which orders you've sent the request to already, Claude Code built it in 5 minutes.
In general, the Amazon Seller UI is a cluster (especially since I have one account for each brand, so I constantly have to switch between them). There are lots of subscription apps to make your Amazon data more useful and accessible, but Claude Code with access to the Amazon APIs pretty much replaces all of them. I spend very little time in the actual Amazon UI now and mostly just ask my trusty assistant for the info that I need.
Just FYI, most of us maintain blacklists of sellers who do that and would never give business to one again, even if it requires paying more. If I bought something from you, that is not permission to email me anything other than a tracking number. Ever. If i like it -- i'll review it mysef. If i dislike it -- i'll email you. Note the direction of comms here.
Making you click a button per-order was perhaps Amazon's way to add friction to this -- to avoid poor users being spammed with endless review requests. I am sad that someone automated the friction away. I hope that one day amazon starts charging sellers a nontrivial fee per such email sent.
The other side of the coin however is a potential decline in indie hacker products.
Fwiw, Google has had a free in-browser tool for this for ages, makes capture really simple on any device with a browser: https://toolbox.googleapps.com/apps/screen_recorder/
Creating tools for your own workflows has become amazing, especially as a creator of anything it feels overwhelming with how many options there are now.
- tell me you haven't heard of OBS studio without telling me you haven't heard of OBS studio. But on a serious note OP, you should really give OBS studio a shot, it is one of the best video recording tools out there, period!
The interesting question is why the OP had bought these subscriptions in the first place if he was happy with less-polished alternatives.
Most people don't make or maintain their own things, period. Vibe coding will mostly cannibalize expensive B2B contracts from a pre-LLM era where integrations and maintenance were expensive.
There might be a lot of people interested in all kinds of small apps for their particular hobbies, interests, jobs, etc
This reminds me of the people who think they can build docsend in a weekend. No, you cannot. You can build a wee throwaway app with some of the features of docsend. But that is not equivalent to what people pay docsend for.
Businesses and SaaS aren’t just a bunch of static code. Code is a part of them, but it’s actually a minority of the work and the service. It’s very common to see founders fall out because CTOs believe product and company = tech and CEOs do not.
If you’re a technical founder, learning this lesson will separate you from the pack.
Intuit does just that and charges north of $100 per month.
Replacing some subscription app like Any.do, Google Calendar, fitness/diet tracking or basically any other CRUD-centric app, needn't be insecure, and a semi-competent developer can easily host it, continue further development (with or without vibe coding) and secure it. There's huge benefit for software developers that do find themselves using many of these apps with active subscriptions to make their own, tailored for themselves, and cut down their spending.
Yes, when it comes to commercialising such software, more work needs to be done (mostly in support and marketing), but for personal use it's fine. The author explicitly states they don't trust vibe coding enough to turn these into products.
The writing is hardly on the wall for all these companies which make little todo list apps and calendars. The vast majority of people could get a LLM to produce an alternative but the lacking they have in basic software engineering would eventually be a hurdle to further development. Most people will continue spending $1.00/month here, and $2.99/month there. There's no reason why software engineers need to do that anymore, unless paying this gives them access to some sort of content repository (music, books) or actual advanced software.
I guess if you're unemployed or in an area with spectacularly low wages, and don't have any ideas of your own that seem monetizable.
>Most people will continue spending $1.00/month here, and $2.99/month there.
If I make 100 dollars an hour as a consultant, and I spend 1 hour to make a local version that never needs any updating on my part to replicate a portion of that functionality I get for $1.00/month, it will take me 101 months to see any profit on my investment of time.
Cut my pay in half and I still need 51 months to see any profit. It would be idiotic for me to waste 1 hour on that.
And let's face it, code when made is a cost center, you will have to keep it up to date (so as to not introduce security hazards etc.) you will never break even much less earn anything for your time.
Something designed without content suggestions, ads, influence and constant un-necessary redesigns, for privacy and to retain their own data.
Good for you economically. Some people are unemployed and underpaid. In fact, most are. Half of your post just came across as you broadcasting your economic success.
The vibecoded threat isn't from Devin, the threat is from 15 year old nerds who no mortgage, partner, or responsibilities. Powered by insecurity, pride, and spite, plus a generous amount of teenage hormones. I would love to see What I could have built if I had Claude back when my reflexes were still powered by youth.
Plus, if you made an app that other people were paying $1/month for. Sell yours to people for $2/month (or $0.50/month, you decide!) and you'd recoup your money much faster than 51 months.
There's another competition, your average folks who end up using physical notebooks, stock Notes/Calendar/... apps and Google Sheets when they figure the subscription isn't worth it anymore / they can't keep up with all of them due to rising CoL.
Three major iterations from now this whole conversation will be quaint and everyone will have always thought this.
Bugs to fix. Dependencies to maintain. Features to add. Data to protect and restore when things go wrong.
You don’t just pay for some code. You’re paying for a service.
The author is very clear that that's not what they are trying to achieve:
"Now, don’t get me wrong, Jabber is not “production quality.” I would never sell it as a product or even recommend it to other people, but it does what I needed from Wispr Flow, and it does exactly the way I want it to."
Whatever the default model was choose for jabber, it just hits my sweet spot: it recognizes my speech faster than typing (I type relatively quickly and speak relatively slowly, so any lag in the recognition can quickly tip the scale).
I honestly don't know what a real production software of this kind would do better (besides the promise that somebody will actually take care of issuing updates for the next time apple breaks backward compat)
If your particular use of a SaaS is not susceptible to security issues (for example you can use it on your local laptop) then a slither of features that are insecure is exactly the thing for you.
Not everyone is going to replace SaaS with half-baked personal implementations, especially the big companies that most SaaS are aimed at, but it will gnaw out some of the long tail of SaaS subscription revenue for sure.
What is a slither in this context? Or should this be "sliver"?
If the article was claiming that you could, you might have a point.
> You can build a wee throwaway app with some of the features of [some commercial app].
Which is precisely and explicitly what the author has done.
There are whole products that succeed by simply being less capable, but cheaper, variations of some other product. This is most obvious in the art space, where you can visually see the reduction in tools and menu items.
This people often don't want support, maintenance, or future development. Their needs have been met.
There's also a story about Emacs at Amazon, where secretaries would code up elaborate things to automate their stuff. It tells you how silly our world is : we're happy with shitty uninteractive systems, languages and OSs, only so that we can fix it with the next version of "genius" AI systems which can barely do anything more than remove pain of dealing with the insane bloat.
A trial, and then a $5 or $10 one time purchase. If ever you're missing some feature that gets added in a later version, then maybe a $5 upgrade.
But why would a simple text or video editor need a monthly subscription?
Why are the options either paid or vibe coded?
Perhaps that’s a worthwhile trade, but you’re still bearing the cost in a different form.
The author cites Loom and a transcription service as their projects. They've just stitched together screen record and a video editor, and made a nice front end for Whisper.
Not diminishing this, quite the opposite. Something I love about coding in the 21st century is that there's so much to build on! A lot of work is about finding two or more cool things that work well together, packaging them up nicely and creating something that's greater than the sum of its parts.
I've mixed feelings about AI, but it's great for this sort of work flow. It's great for writing the tedious glue/GUI around established libraries.
Will have to find ways of providing value outside of technology, as the technology parts, AI not included (it will cost money), will be free. We'll probably need to sell some sort of human connection/personality or tangible goods to make money in apps.
Incidentally, I ran into something like this with WhisperFlow last year. Used it for a few weeks, loved it, basically hardly typed for the month and just spoke to my system/terminal etc.
But, I ran into a unique challenge. Barking orders at my computed for 8 hours a day made me realize that I was changing how I communicated with people. Being nicer was easier to solve, but speech-to-text made me less articulate. I wasn't very articulate to begin with -- which is something that I have wanted to solve for a while.
So I built my own STT app, that works in a similar way as whisperflow, with a few notable exceptions. Minor: it has dictionaries, snippets etc on a per app/website basis. Major: most notably it has rubrics on how I want to communicate in different contexts, ex: Biz Exec over email, Principle engineer in my ide/terminal. etc
And scores me on areas like conciseness, flow, logical flow/ease to follow, clarity etc. every time I say anything. 10 weeks in I'm noticeably more articulate than I've ever been.
Some software essentially maintains this model, and I think it's fine. If something is going to charge monthly I expect it to either be much cheaper, or have some web service that they need to keep up.
Most of the paid subs is specifically because i can’t easily do it myself. ie there is an attribute there that extends beyond coding skill. I can’t easily defend myself against Ddos or can’t easily do multi region redundancy etc. Stuff that even a nice home server and coding plan just can’t cover
I see basically no reason why Salesforce can't be taken down by a 2-3 person team right now. Shopify? Everything Intuit? Atlassian (Jira, Confluence)?
The IKEA effect does the rest
I also run a Hugo blog, running on a very old version. I'll be honest, I'm tired of editing it in VS Code with the Markdown Rendering on the side.
I was looking for something new, completely separate. Really hoping this project will work for my needs too. Thank you so much!
kylecazar•2w ago
It used to be that you offer subscriptions only if there are ongoing costs, and a one-time payment if not (utilities, local, etc). SaaS kinda ruined that.
I'd welcome a boom in DIY vibe-coded utilities for personal use.
mr_mitm•2w ago
theshrike79•2w ago
But I'm not going to start adding features to my bespoke utility to fix someone else's problem.
pixl97•2w ago
Sometimes you just want an app does X and Y, but not A, B and Z.
TheGRS•2w ago
I also don't think any particular idea is off limits for making a profit, if you do something and you do it well, you can charge a fee. But if the free hobby version is better then you best find a way to justify the price.
quesomaster9000•2w ago
The most recent example is I wanted a simple home security system with presence detection and a private control panel, none of the free ones hit my requirements, or require custom hardware, or lock you into a cloud, or assume you can spin-up some containers - or are super enterprise grade stuff.
Within about 2 days I had an android app for my tablet, Google FMDN integration, fingerprinting of my other devices, all controllable via Telegram from any of my phones with alerts that "just work" wherever I am and include an inline gif snapshot.
What I wanted didn't really exist as any individual product, so I absolutely see the appeal of DIY vibe-coded stuff, and a day of the build time was optimizing the OpenGL motion-detection pipeline with shaders & DMA which in itself was good to learn about.
sevenseacat•2w ago
carlosjobim•2w ago
No way. Maybe most Linux users. MacOS, iOS, Android, and probably Windows users are all app-first.
sevenseacat•2w ago