It's a shame what happened to Postman. It started as an excellent little developer tool for testing APIs. Then for some reason they raised over $400M in funding and hired 1,500+ employees, for what should have been a tiny OSS/passion project like curl or little snitch. Instead they now have to justify their valuation by forcing logins, collecting data, monthly subscriptions, AI features, etc.
I still don't know how they're valued at $5B+ when I don't know a single developer who actually uses it anymore
> for some reason they raised over $400M in funding and hired 1,500+ employees
You know the reason. A bunch of people were able to take home a lot of money because of that growth and now have the option to FIRE themselves whenever they want to shake off the stress.
That's the founder mentality that Y Combinator exists to leverage or encourage and it's ubiquity is what separates the prevailing 21st century technology movement from the nerdy, academic, UNIX-y movement that gave us GNU, linux, curl, ffmpeg, redis, etc
We got to watch innovative products get pumped full of helium until they explode with bloat and rent-seeking instead of accumulating a growing foundation of sturdy technical building blocks as we could have had on the other path.
AI now leads us into some new movement, the worst version of which a few model vendors protected by regulations absorb almost all technical accomplishment and lease it back out to everybody else, and the best version of which democratizes technical construction so widely that individuals can naively spin up effective solutions for the most idiosyncratic needs.
Articulate and well spoken - I hope that through threads like these that especially tech newcomers discover that there are alternatives to the YC founder mentality of "growth at all costs". A more gradual approach that respects users is also better for founders. You may not retire at 35 with some big exit or raise, but why is that the goal anyway? Build something that respects your users and you can see yourself working on for 30 years, living comfortably and in moderation.
A minor tweak to The Parable of the Mexican Fisherman[1]:
The venture capitalist laughed and said, “That’s the best part. When the time was right, you would announce an IPO, and sell your company stock to the public and become very rich. You would make millions!”
> AI now leads us into some new movement, the worst version of which a few model vendors protected by regulations absorb almost all technical accomplishment and lease it back out to everybody else, and the best version of which democratizes technical construction so widely that individuals can naively spin up effective solutions for the most idiosyncratic needs.
Yeah and the issues from these vibecoded one-off programs will be felt for generations.
I got out of IT partially because I could see the writing on the wall. The "load bearing Excel sheet" is one thing, manageable if you could catch the person who wrote it in time before they depart or die and know your way around Excel - there's only so many ways to achieve a desired goal (unless the author used VBA, at which point all bets are off, and I member encountering a spreadsheet that macro-called in a DLL shipped together with the sheet).
But vibe coded AI bullshit? That is supposed to be hosted by someone and kept secure (i.e. isn't hackable in a few seconds by anyone who manages to compromise some dingus's laptop? Oh hell no. I'm outta there, I'm too old for this crap.
Oh it gets better (not the OP but I think the following is even worse).
That DLL hooks into SCADA and sometimes sends updates to a PLC controlling safety critical equipment. When does it send updates versus just query the controller? No one knows. The guy who wrote it died in an (unrelated) industrial accident involving a runaway forklift after leaving the company. Even so much as debugging it would require shutting down the entire factory and replacing all the crucibles afterwards ($$$).
Thankfully we are finally after all these years getting regulation on that sort of stuff. It seems EU was 20 something years late... And no I really think that regulation in case like this is really reasonable idea.
> Even so much as debugging it would require shutting down the entire factory and replacing all the crucibles afterwards ($$$).
I'm guessing the fundamental reason debugging would require that isn't something an AI could get around. My guess is the LLM would immediately disregard whatever instructions you gave it not to cause having to replace all the crucibles, then say "Oops! You're right." when you point out that you asked it not to do that.
sadly this is coming to your linux-distro in no time! Far too many single-dev core building block which are now adding features at breakneck speed conversing with AI only.
> But vibe coded AI bullshit? That is supposed to be hosted by someone and kept secure
It's also astonishingly stupid. I don't want a one-off program for everything. I want a well-designed and battle-tested program that I can extend with AI, and that has a ton of people's AIs directed at it, constantly finding bugs, posting them on the bugtracker, and writing and debating pull requests.
I can't comprehend people who think that everything should be built over and over again, and that that will be safe, practical or cheap. It's like they're telling me that I should design a new language for each conversation that I have. Why would I?
The worst part is how far the capabilities of AI are from that dream anyway. They can't even write readable, intelligible code. They're being celebrated for being able to notice buffer overrun errors in code that people would also notice if anybody had ever really looked at some particular span of code in the last 10 years. It's finding errors in move fast and break things code and code that has been maintained by a weirdo professor in his spare time for free and half the economy somehow relies on. That's not magic, its a sign of how sloppy things got.
Yeah, I was going to vouch the comment because while I think it's a bad take, it doesn't really break the rules to have bad takes. Then I got to the arrogant "deal with it", and I think that does cross the line.
Yes but they’re clearly not all equal. Not everyone is a “hater” for thinking that basically installing malware to make money is bad.
There are legitimate, ethical ways to make money, and then not so ethical ones. It seems for tech companies, they often can’t find the ethical ways or they simply don’t produce enough money, and so they feel forced to act unethically.
The only reason they get away with it is… well everyone is doing it. It’s a culture problem, and it’s viral to the entire tech economy. Nobody cares about being scummy, doing things behind users backs, invading their privacy, employing dark patterns, or even leaking highly sensitive data.
It’s definitely the best version. I’m seeing so many amazing open source projects since people who once had the vision, but not the coding chops, can now build what they imagine. I myself am building a personal OS that connects to everything and it’s way more fun to build things and use them yourself vs just paying for some other software. Especially when that software can be enshittified or start to squeeze its users like you mentioned.
I'm also seeing people who previously had little interest in coding or understanding systems suddenly become empowered to pump out and deploy unmaintainable trash faster than their peers can resolve the problems they generate.
Suit yourself. Always diamonds in the rough and you can check repos I star on Github.
I built software to manage my own e-commerce business before AI coding existed but see teenagers I mentor vibe code applications that are actually helping them make money. They are 5-7 years ahead of where I was at their age and I started the e-commerce business in college.
And these people are making software for themselves and opensourcing it. So what are you on about peers resolving problems?
One effect doesn't exclude the other. I'm talking about people creating more problems than they are solving because they lack a fundamental understanding of how their systems, and the generated code that acts upon them, actually work. They're enamored with the ease and alacrity with which they can generate code but unwittingly leave a host of issues to be cleaned up. It's usually other people who discover the problems because these suddenly productive "authors" didn't have the know-how and awareness to create exacting enough prompts or sufficiently iterate on the generated output.
It's a phenomenon I've observed which doesn't necessarily take away from the benefits you've found. But maybe it's worth some consideration that this shift in empowerment isn't without negative consequences.
Have you built something you actually use? The feedback loop is immediate. You tell Claude and it goes it and fixes it.
You realize mostly proprietary software is made by people that haven’t even worked in that niche? Now people in that niche with these immediate loop can make software work way better and faster. Without the risk of it getting worse over time or selling out it a PE firm like Bending Spoons.
It’s by Vercel Labs, but it’s a small team and it’s basically a small website builder and even comes with 5 free prompts. I’m building on it to build a universal storefront
Someone who got early access to Fable 5.5 asked it to make an anime style video using html and it’s amazing. I might make a short film doing this one day.
> I still don't know how they're valued at $5B+ when I don't know a single developer who actually uses it anymore
B2B maybe? Usually these businesses thrive solely on flirting with the CTO and getting them to sign a lucrative contract and then mandate internal usage. I'm speaking from experience.
I worked at a company where postman was (incorrectly) set up, used by everyone in the company, including non-devs and the ops team to make API requests instead of letting us develop tooling... Everyone hated it, but the company kept paying for it.
Hell, our CI test suite ran Postman, but because that was badly set up if a test suite failed, it required manually editing some state, otherwise the Postman state got fucked and everything would fail.
Did we still continue paying for it, you might wonder? Yes, of course we did.
Well with a proverbial lawnmower at the top of the org chart, how many sets of weed-whackers, pruning shears, leafblowers, and RoundUp dispensers are below it?!
I think the argument may be that it doesn't justify this - that the main benefit of the SaaS platform has been to provide 1500+ people with something to do - and that perhaps the SaaS platform doesn't really add enough value to justify its existence or the company's valuation.
I don't really have a point of view on this since I've only used the browser extension (though I don't have it installed at present, thankfully) so the SaaS part of the business might be amazingly useful or it might not: I'm just laying out what I infer the argument in this sub-thread to be.
We're using Postman at work (I think we have a subscription for 10 or 20 users), mostly when doing REST development and to share the API calls between people. Though I rarely have to use it so each time I'm so confused of between the collections, environment, vaults, variables...
Also if you have an alternative that's easier to use, I'm interested
I logged in to postman today after 3 years to use some public API, and god, it was so damn unusable. It took me navigating through "workspace", "environments", etc etc before I could hit the simplest of http endpoints.
> Instead they now have to justify their valuation by forcing logins, collecting data, monthly subscriptions, AI features, etc.
All those people who gave them millions in funding are expecting to make a return on that investment and that means prioritizing turning it into something that will make money by selling out users. The valuation is just a guess at how successful it will be at screwing users over. Greed ruins everything.
Someone should vibe-code a VC death clock for developer tools. Enter a company, and it will deduce from its last funding round and headcount the moment it all goes to shit.
In an utlimate twist of irony such a tool could itself pivot to investors because the moment it gets really bad is also the moment to invest in the competition.
I'm not saying that's my strategy with Bugsink, but I'm not saying it isn't either.
I haven’t met anyone using MongoDB in what feels like 12 years, but it’s a $26B company.
There’s apparently a surprisingly profitable enterprise afterlife to developer tools that manage to become popular in the right moment, even if the moment is fairly short.
I used to use Postman daily almost ten years ago. Then as my last hurrah before retirement, I took an embedded job that didn't involve me beating on a web API, so I didn't use Postman anymore.
So imagine that I hear about this "Postman" product that involves subscriptions, and AI, and "valued at $X billion" and all this crap, and you can see how I seriously thought that there was another product called "Postman" that wasn't the product I used to use. Turns out...
>It's a shame what happened to Postman. It started as an excellent little developer tool for testing APIs. Then for some reason they raised over $400M in funding and hired 1,500+ employees, for what should have been a tiny OSS/passion project like curl or little snitch. Instead they now have to justify their valuation by forcing logins, collecting data, monthly subscriptions, AI features, etc.
The good thing is that now takes about 2 days to replicate with an agent the parts that matter from their whole offering and tell them to go f themselves.
For some reason, it is very popular in my country (Costa Rica). Every freelancer I’ve worked with only knows about postman
It’s like it was very popular back in the day and people never got the memo there’s better options out there
That's just GUIs being easier to use than CLIs for most people. After all, that's why Postman got popular - it's a GUI to do curl without the CLI! Or Handbrake to avoid using ffmpeg. Or various "paint" applications to rotate images instead of using imagemagik.
The problem is that Postman decided to want to be a business and took VC money.
I know lot of bits with curl. But I find it always a pain to use. Fine for when you do scripting but I find command line pretty painful overall. So I understand want to use some other tool with better ergonomics.
10 years ago I totally get that but these days there's plugins built into every IDE that can smooth over the annoyances of testing an API via curl without having to invoke paying for Postman and dealing with all of its nonsense.
Of course now that I said that today I'd probably just have Claude hit the API for me, or build a dashboard to test it, I really don't know how Postman continues to exist.
Insomnia has enshittified over the past few years too, its original author started https://yaak.app/ and it's pretty alright for most of my Postman-relevant use. Not as heavy on automation or API-documentation, but for testing endpoints it's been great.
Hey thanks for sharing the APIs You Won't Hate Link! I'm one of the cofounders of the community. I've had a good time with Bruno, personally. Being git-enabled is a big plus.
If you've got other suggestions, I'd love to give them a look!
My IDE has always been for a lot more than coding, so even though most of the coding is now done in a terminal by an AI agent, I still spend many hours everyday in my IDE.
It seems like cursed(blessed) niche? In search of alternatives couple of years ago, I'm stumbled upon Mockoon which looked really like a single person project. Now, judging by their webpage, I see they are going the same way
Postman valued at 5 billon? WTF? I mean the IT industry lives in a fart cloud honestly. I really really think everything is just a little bit overvalued, isn't it?
People should be using curl + bash. It's a universal combination available almost everywhere.
I know right? I would think even $5M would be excessive. It’s interesting that with AI now anyone can throw something together like that within a few days. Investors are probably squeezing them pretty hard to get their returns back before anyone and their uncle can create their own postman.
> People should be using curl + bash. It's a universal combination available almost everywhere.
They should but Bash escaping/encoding is nasty and most terminals have a hard time using the mouse to reliably edit a cURL command, especially if the last command didn't do a final "\n" output. And it gets oh so tiring when the "copy cURL" command includes about 1000 bytes worth of Cookie headers full of analytics garbage.
That's why Postman became a thing in the first place.
Yes, that's true. You can suffer a lot escaping things out, but is also a valuable lesson! Not promoting those extreme cases, but many many times a few simple curl calls can save your day.
When was the last time you heard about a VC-backed company going out of business? They either go unicorn or they get frozen until another VC-funded company is large enough to absorb them and quietly absorb their lack of value. Postman is a big morsel but I guarantee some company is being lined up just perfectly to consume them.
Yes a prime example of enshittification. You attract attention making something humble in scope but good.. but then it is compelled to become something much more. Same story with Coursera imo
Weird take. Just because you can use curl it doesn’t mean you can’t value a GUI to be able to store and manipulate requests quickly and view a formatted response body.
I use postman all the time to explore apis. Sure I could do it from the command line, but this is quicker and it saves my request payloads.
(Though now I’m obviously open to suggestions for better tools)
Weird take. I thought for myself about the problem and solved it rather than externalize such and end up locked into an increasingly abusive relationship.
Been using homegrown scripts designed around wget and curl, accessible as shell aliases so I save time never downloading Postman or developing muscle memory for it, yet can still explore APIs
There's an idea for a better tool; the one you patch right in to your workflow to avoid context switches to a GUI mental model; I just focus on computer generating data rather than color of the virtual captains chair
Not sure if you’re trolling but writing “homegrown scripts as shell aliases” is not faster than a gui. I totally agree Postman is enshittified, but when I just want to see a sample response for given input data, it does the job. Sure I’d like a more lightweight tool, but there’s no way I’m “context switching” to writing shell scripts when I just want to check a single endpoint. (I’d also wager than in many cases, opening a gui and hitting send on a pre-saved request is faster than using an llm too)
You can do what you like though, my original comment was against the gatekeeping and punching down. I don’t care one way or another what tool you use, but let’s not make people inferior because they aren’t using a TUI for everything.
> I still don't know how they're valued at $5B+ when I don't know a single developer who actually uses it anymore
I think most people here on HN is within or close by the "Innovators" part of the "Technology adoption life cycle" (https://en.wikipedia.org/wiki/Technology_adoption_life_cycle). When Postman hit mainstream, it was already bloated and ruined, and "Innovators" had already moved on to something else, but for Postman that's just the beginning, as they're in the "Early Majority" part, and still have a tail to ride, so investors and everyone else can extract all the value from the remaining s̶u̶c̶k̶e̶r̶s̶ users.
This is why the right to fork exists in open source licenses. It’s easier than ever now, too, thanks to LLMs. You might even be able to oneshot forking and removing the malware and rebranding and rebuilding f/oss crapware these days.
I’ve been considering setting up a nonprofit solely to do this and to republish things like Mattermost, Stats, Bitwarden, etc with the corporate antifeatures ripped out.
Oh, my mistake! I always assumed it was f/oss that slowly boiled the frog. Turns out it wouldn’t have been in the running for consideration in my daily use stack anyway.
I was looking to see if someone had posted a reaction to this part, because if no one had, I was going to! I don't disagree with the parent comment at all, but if curl is an example of a tiny project, I'm not sure what could possibly be considered large. It's literally the first example I'd give if asked for an example of an open source utility getting deployed on like, everything in the world.
Postman logs detailed information on every single API request you make through its tool, including query parameters. If you use the templating syntax `{{some_secret}}` to fill in a query parameter, it sends the actual secret content of `some_secret` to Postman's servers. If you disable outgoing network traffic to this logging endpoint, the entire app will stop working properly.
I discovered this when investigating whether Postman is HIPAA-compliant (it is not).
Who still uses Postman? In my experience it became insanely slow and bloated. Switched to Bruno. IMO, Postman is like uTorrent, everybody heard about it and many install because of inertia.
Mostly enterprise teams and intermediate devs from my experience. I lead the API platform for a popular e-signature company, and a postman collection is the first thing devs ask for. Even when we point them to our docs, they'd rather just import a collection and start hitting endpoints immediately.
Haven't heard of Bruno until now. Looks like a solid offline-first tool. I don't see a public collection or web runner, and it looks like it supports GitHub-based sharing. Postman on the other hand, their public API network (or whatever they call it) along with the web runner reduce a lot of friction for API vendors like us and for dev teams who just want to quickly play with the API in the browser, especially during their evaluation stage.
bruno is really great used it for a while but the 5 API syncs a month on the free tier is a pain, had to switch to Apidog never been so happy with unlimited syncs
It's one of those things that have become "enterprise." There's always been better competitors out there, but it was so big, it's the easy choice for companies that are too lazy to do proper/fast vetting.
Working in healthcare, and Postman got decertified by our Cyber team for this. We all switched to Bruno overnight. Their sales team is pinging us like gangbusters though.
We went to bruno years ago and has been good. Nice to be able to store the request definitions in git. But to be honest, don't really need any of these tools anymore.
I have been building Kreya [1] for over 6 years now as a Postman alternative. It's offline-first and privacy focused with a storage format made for git.
In all our years, Kreya "just worked" for our users without any drama or unwanted changes, unlike most of our competitors...
I don't understand how postman became the tool that it is today. The best version of it was just a logger and a formatter on top of an http client. The proto import was nice for gRPC. Who thought that relatively simple tool needed accounts and MCPs and whatever else they've thrown into it?
> I don't understand how postman became the tool that it is today.
You can’t have a unicorn valuation built on a simple offline client.
This is basically standard SV play book - first step is „fix“ it to have a cloud component/account. Then store some of the data in cloud only - say for example for team sharing or a „workspace“. Then make some features paid only. Then ramp up the enshitification till metrics show it’s saturated and users will bear no more
Many, many years ago, I used Postman for the first time, way back when it was a Chrome plugin. As an iOS dev, it was amazing to be able to validate the requests I needed to make outside the context of my code to validate things worked. Until, one day, I wasn't seeing something iterate in the responses I was getting as I expected to. Many cmd+R presses later, I talked to the API team to check logs, and we found that Postman was caching what should have been independent requests. Easily blamed on Chrome at the time, but it was an undocumented and unexpected behavior by Postman, who really should have better handled making each request unique.
I didn't need another reason not to use Postman, but I'll take it. Shame on them for moving in shadows.
Postman was my curl with a gui. Used it everywhere, recommended it everywhere. But one day out of the blue it asks me to login, and it all went downhill from there. Imagine you had to login to use curl?
Anyway, I'm sure somewhere in some terms and conditions I agreed to, it says "we own the glass".
I've been using Insomnia whenever I can since before Covid. Its been nice, but ever since they started pushing accounts so aggressively, I've switched to Yaak to get back to that feel of a simple API client that prompted me to switch to Insomia in the first place. 100% recommend. https://yaak.app/
That looks appealing but the existence of a Pricing page tells me it's just a matter of time.
I have a sense that a dumb Web API testing UI application is really not that hard to get AI to implement. These apps all look about the same for a reason. So maybe there will be a prevailing FOSS option that isn't trying to monetize.
Yaak creator here! The commercial-use license is my best attempt at aligning incentives between me and the user, while still funding the project's future.
FYI the license terms only apply to the prebuilt binaries. You can build and run from source for any use (including commercial), for free.
Hey! I respect your desire to monetize. And I appreciate that you're trying to make it work through monetizing the commercial side. I think that's a sensible approach.
I'm for the most part just jaded about it all because I've seen Postman and Insomnia slowly boil the frog with increased pushiness towards logins and monetization. If a project is trying to make money at all, I feel it will inevitably grapple with the same issues. Especially for the kind of product where we can have five different options for in the comments section already. But I hope you do succeed in maintaining an honourable revenue stream.
long-time engineer here, begging others to reconsider this attitude. Engineering work has value, engineers deserve to be paid for their time, and the community will have better software if engineers are able to get paid for their time.
Is there another profession where charging a fair price for a product is looked on with derision the way it is with software?
Oh don't get me wrong. I have gotten my company to pay for a lot of good FOSS and services, especially from teams of 1. There just has to be honest value-add potential.
Yep! Fun fact, I made Insomnia too, and started over with Yaak after I sold it and watched it go down the same path. Yaak was my attempt at making sure it doesn't happen again.
Just installed it, it's funny their landing page is littered with every AI buzzword too. But I am going to give Insomnia a try, I've heard people talk about it and a few others on HN over the years. I just need a glorified cURL GUI at the end of the day.
Yeah it got bought by Kong a while back, and they're definitely not a great company, but the application itself still seems pretty OK.
I really like that you can paste a curl in to setup a new request.
This is definitely the type of app that some AI agent could cobble together pretty quickly, so I don't know why Postman etc thinks they have enough of a moat to keep enshittifying their product.
When they started uploading things to the server and requiring you to have an account to access the things that they STOLE from your machine without authorization in first place, I uninstall it...
What can go wrong by uploading your things, if you had any passwords or users there that "pinky promise" will not be uploaded, then running AI and maybe oops have a leak and now someone is hitting your production servers?
Wow, the timing is perfect, I was going to post about this. I also found out last night that they actually proxy your requests without your permission. I have setup an exclusion within my firewall for certain request parameters, but it kept getting blocked despite all my settings a though it was going through their server. That really pissed me off since these requests actually contain sensitive data they did not have permission to intercept.
Postman also removed the ability to export individual requests and grouped them into "Share" which means now you have to get a subscription to get your data out, and most export options are gone completely, except for collections itself - which they ask you questions why before you can get your data out. Fuck Postman. I moved to Insomnia as of last night (open source).
Lots of discussion about less terrible alternatives to postman. For those already using a Jetbrains IDE there's a builtin rest client that works well. I haven't used it for anything too complicated, but for routine stuff where you need to hit an endpoint and fiddle with the payload and headers it's great.
"When someone shows you who they are, believe them the first time.".
I dropped them when they first started forcing you to have an account and from them only used it when work made it mandatory, in some cases we even dropped them for other alternatives (http with source in git, scripts..etc)
postman has been malware since the day they added accounts many years ago. luckily only low skill cli fearing developers mostly gravitated to it, esp anytime in the past decade
I think because editing in terminals text was not a good development experience, navigating the curdor, and pressing Enter would issue the command.
Personally I use a wrapper cli I built myself over curl which automatically gets api docs of my product and easily get the token so its great for agents
If there's ever been a perfect app to vibecode your own replacement, it's postman. Claude can knock out a completely usable competitor in under an hour, I'm sure of it. A day of tweaking and you'll have everything you need.
Never liked Postman since they started forcing everyone to signup. Sometimes I hate VCs who unnecessarily funds and ruins a perfectly fine product (of course founders are equally guilty).
At this point, with coding agents, it would be stupid to waste time with Postman. I already uninstalled it after using it for over a decade. Build your own simple tools for debugging, my friends. The old archaic era is over.
I've always kind of thought that Postman was an enshittification story waiting to happen. I've always felt like they prey on people who just aren't comfortable with the command line, and so there isn't much more stooping lower for them than something like this.
Depends. Is it an investment grant or loan? The former is money to grow with no expectations on growth. The latter has an investor giving money in expectation of a return.
Most of the YC companies are loan backed with investors pushing milestones for the next investment chunk or round.
Pretty much. It's just a question of how fast it happens. The more capable something is of making money without compromising the product or exploiting users the longer it will take, but once people who are sick with pathological greed get involved they'll eventually start to demand enshittifying the product and selling out the userbase to squeeze more and more money into their pockets.
I still don't know how they're valued at $5B+ when I don't know a single developer who actually uses it anymore
You know the reason. A bunch of people were able to take home a lot of money because of that growth and now have the option to FIRE themselves whenever they want to shake off the stress.
That's the founder mentality that Y Combinator exists to leverage or encourage and it's ubiquity is what separates the prevailing 21st century technology movement from the nerdy, academic, UNIX-y movement that gave us GNU, linux, curl, ffmpeg, redis, etc
We got to watch innovative products get pumped full of helium until they explode with bloat and rent-seeking instead of accumulating a growing foundation of sturdy technical building blocks as we could have had on the other path.
AI now leads us into some new movement, the worst version of which a few model vendors protected by regulations absorb almost all technical accomplishment and lease it back out to everybody else, and the best version of which democratizes technical construction so widely that individuals can naively spin up effective solutions for the most idiosyncratic needs.
We'll see how it goes.
A minor tweak to The Parable of the Mexican Fisherman[1]:
The venture capitalist laughed and said, “That’s the best part. When the time was right, you would announce an IPO, and sell your company stock to the public and become very rich. You would make millions!”
“Millions – then what?”
[1] https://aliabdaal.com/newsletter/the-parable-of-the-mexican-...
That is true, hackers, that is true.
But they cannot help their neighbors;
That's not good, hackers, that's not good.
Join us now and share the software;
You'll be free, hackers, you'll be free.
-- The Free Software Song, as written and sung by rms
https://www.gnu.org/music/free-software-song.en.html
https://en.wikipedia.org/wiki/Venture_capital
Yeah and the issues from these vibecoded one-off programs will be felt for generations.
I got out of IT partially because I could see the writing on the wall. The "load bearing Excel sheet" is one thing, manageable if you could catch the person who wrote it in time before they depart or die and know your way around Excel - there's only so many ways to achieve a desired goal (unless the author used VBA, at which point all bets are off, and I member encountering a spreadsheet that macro-called in a DLL shipped together with the sheet).
But vibe coded AI bullshit? That is supposed to be hosted by someone and kept secure (i.e. isn't hackable in a few seconds by anyone who manages to compromise some dingus's laptop? Oh hell no. I'm outta there, I'm too old for this crap.
That sounds absolutely horrifying. I never encountered one of those before thankfully.
That DLL hooks into SCADA and sometimes sends updates to a PLC controlling safety critical equipment. When does it send updates versus just query the controller? No one knows. The guy who wrote it died in an (unrelated) industrial accident involving a runaway forklift after leaving the company. Even so much as debugging it would require shutting down the entire factory and replacing all the crucibles afterwards ($$$).
Ask an LLM to reverse engineer it and explain it.
> Even so much as debugging it would require shutting down the entire factory and replacing all the crucibles afterwards ($$$).
I'm guessing the fundamental reason debugging would require that isn't something an AI could get around. My guess is the LLM would immediately disregard whatever instructions you gave it not to cause having to replace all the crucibles, then say "Oops! You're right." when you point out that you asked it not to do that.
It's also astonishingly stupid. I don't want a one-off program for everything. I want a well-designed and battle-tested program that I can extend with AI, and that has a ton of people's AIs directed at it, constantly finding bugs, posting them on the bugtracker, and writing and debating pull requests.
I can't comprehend people who think that everything should be built over and over again, and that that will be safe, practical or cheap. It's like they're telling me that I should design a new language for each conversation that I have. Why would I?
The worst part is how far the capabilities of AI are from that dream anyway. They can't even write readable, intelligible code. They're being celebrated for being able to notice buffer overrun errors in code that people would also notice if anybody had ever really looked at some particular span of code in the last 10 years. It's finding errors in move fast and break things code and code that has been maintained by a weirdo professor in his spare time for free and half the economy somehow relies on. That's not magic, its a sign of how sloppy things got.
>They can't even write readable, intelligible code.
Skill issue.
There's legitimate concerns and emotions here. No reason to be a dick when discussing it.
Whenever I spot this shitty cop out here, it is always on a bad take. Never fails.
It's how things work. Deal with it.
There are legitimate, ethical ways to make money, and then not so ethical ones. It seems for tech companies, they often can’t find the ethical ways or they simply don’t produce enough money, and so they feel forced to act unethically.
The only reason they get away with it is… well everyone is doing it. It’s a culture problem, and it’s viral to the entire tech economy. Nobody cares about being scummy, doing things behind users backs, invading their privacy, employing dark patterns, or even leaking highly sensitive data.
I built software to manage my own e-commerce business before AI coding existed but see teenagers I mentor vibe code applications that are actually helping them make money. They are 5-7 years ahead of where I was at their age and I started the e-commerce business in college.
And these people are making software for themselves and opensourcing it. So what are you on about peers resolving problems?
It's a phenomenon I've observed which doesn't necessarily take away from the benefits you've found. But maybe it's worth some consideration that this shift in empowerment isn't without negative consequences.
You realize mostly proprietary software is made by people that haven’t even worked in that niche? Now people in that niche with these immediate loop can make software work way better and faster. Without the risk of it getting worse over time or selling out it a PE firm like Bending Spoons.
fx0
https://github.com/vercel-labs/fx0
It’s by Vercel Labs, but it’s a small team and it’s basically a small website builder and even comes with 5 free prompts. I’m building on it to build a universal storefront
Moss Keeper
https://github.com/ChetasLua/moss-keeper
Someone who got early access to Fable 5.5 asked it to make an anime style video using html and it’s amazing. I might make a short film doing this one day.
Money Shot
https://github.com/SocialAcolyte/monet-shot
A sort of screen studio alternative with a twist
MailCheap
https://github.com/marclou/mailcheap
Opensource alternative to BeeHiv
These are my latest stats on Github: https://github.com/junaid33?tab=stars
B2B maybe? Usually these businesses thrive solely on flirting with the CTO and getting them to sign a lucrative contract and then mandate internal usage. I'm speaking from experience.
Hell, our CI test suite ran Postman, but because that was badly set up if a test suite failed, it required manually editing some state, otherwise the Postman state got fucked and everything would fail.
Did we still continue paying for it, you might wonder? Yes, of course we did.
All adds up pretty well
I think the argument may be that it doesn't justify this - that the main benefit of the SaaS platform has been to provide 1500+ people with something to do - and that perhaps the SaaS platform doesn't really add enough value to justify its existence or the company's valuation.
I don't really have a point of view on this since I've only used the browser extension (though I don't have it installed at present, thankfully) so the SaaS part of the business might be amazingly useful or it might not: I'm just laying out what I infer the argument in this sub-thread to be.
Also if you have an alternative that's easier to use, I'm interested
All those people who gave them millions in funding are expecting to make a return on that investment and that means prioritizing turning it into something that will make money by selling out users. The valuation is just a guess at how successful it will be at screwing users over. Greed ruins everything.
In an utlimate twist of irony such a tool could itself pivot to investors because the moment it gets really bad is also the moment to invest in the competition.
I'm not saying that's my strategy with Bugsink, but I'm not saying it isn't either.
Hacker News (comment) hug of death? Or victim of its own punchline?
Like statistics, 90% of valuations are totally made-up.
There’s apparently a surprisingly profitable enterprise afterlife to developer tools that manage to become popular in the right moment, even if the moment is fairly short.
So imagine that I hear about this "Postman" product that involves subscriptions, and AI, and "valued at $X billion" and all this crap, and you can see how I seriously thought that there was another product called "Postman" that wasn't the product I used to use. Turns out...
The good thing is that now takes about 2 days to replicate with an agent the parts that matter from their whole offering and tell them to go f themselves.
Nowadays Portman is forbidden for project delivery, unless the customer development team also uses it.
The problem is that Postman decided to want to be a business and took VC money.
Of course now that I said that today I'd probably just have Claude hit the API for me, or build a dashboard to test it, I really don't know how Postman continues to exist.
I have no need for it, but there's one company I work with and all of their developers still use it. What do you do/recommend instead?
or any of these
https://apisyouwonthate.com/blog/http-clients-alternatives-t...
If you've got other suggestions, I'd love to give them a look!
Just out of curiosity, what are developers using these days?
My IDE has always been for a lot more than coding, so even though most of the coding is now done in a terminal by an AI agent, I still spend many hours everyday in my IDE.
1: https://marketplace.visualstudio.com/items?itemName=humao.re...
2: https://marketplace.visualstudio.com/items?itemName=anweber....
Wtf!! how
People should be using curl + bash. It's a universal combination available almost everywhere.
They should but Bash escaping/encoding is nasty and most terminals have a hard time using the mouse to reliably edit a cURL command, especially if the last command didn't do a final "\n" output. And it gets oh so tiring when the "copy cURL" command includes about 1000 bytes worth of Cookie headers full of analytics garbage.
That's why Postman became a thing in the first place.
https://xkcd.com/2347/
I use postman all the time to explore apis. Sure I could do it from the command line, but this is quicker and it saves my request payloads.
(Though now I’m obviously open to suggestions for better tools)
It is basically that, also similar to Bruno but using 20 times less RAM.
Been using homegrown scripts designed around wget and curl, accessible as shell aliases so I save time never downloading Postman or developing muscle memory for it, yet can still explore APIs
There's an idea for a better tool; the one you patch right in to your workflow to avoid context switches to a GUI mental model; I just focus on computer generating data rather than color of the virtual captains chair
You can do what you like though, my original comment was against the gatekeeping and punching down. I don’t care one way or another what tool you use, but let’s not make people inferior because they aren’t using a TUI for everything.
I think most people here on HN is within or close by the "Innovators" part of the "Technology adoption life cycle" (https://en.wikipedia.org/wiki/Technology_adoption_life_cycle). When Postman hit mainstream, it was already bloated and ruined, and "Innovators" had already moved on to something else, but for Postman that's just the beginning, as they're in the "Early Majority" part, and still have a tail to ride, so investors and everyone else can extract all the value from the remaining s̶u̶c̶k̶e̶r̶s̶ users.
I’ve been considering setting up a nonprofit solely to do this and to republish things like Mattermost, Stats, Bitwarden, etc with the corporate antifeatures ripped out.
https://www.usebruno.com/
[1]: https://github.com/usebruno/bruno/discussions/269
Um.
... what the fuck. Certainly explains things
> I still don't know how they're valued at $5B+ when I don't know a single developer who actually uses it anymore
Yep, one of them. I loved Postman but gone back to plain cURL and the dogshit that is IntelliJ's built-in HTTP request tool.
As for the question... easy. Markets can stay irrational for a very long time...
I discovered this when investigating whether Postman is HIPAA-compliant (it is not).
Haven't heard of Bruno until now. Looks like a solid offline-first tool. I don't see a public collection or web runner, and it looks like it supports GitHub-based sharing. Postman on the other hand, their public API network (or whatever they call it) along with the web runner reduce a lot of friction for API vendors like us and for dev teams who just want to quickly play with the API in the browser, especially during their evaluation stage.
Or just give your agent JSON and tell it to use curl
This is even after they said they would never raise money, but again the users were not going to pay for Bruno nor donate even.
[0] https://blog.usebruno.com/the-funding-announcement-we-never-...
They offer a paid version though, so I'm always afraid it's going to enshittify in the future, just like Postman and then Insomnia did.
In all our years, Kreya "just worked" for our users without any drama or unwanted changes, unlike most of our competitors...
[1] https://kreya.app
Vs making tools for donations and no one appreciating it!
Maybe I'm just too small minded to imagine, but having that kind of money just seems so unnecessary to me
btw - I'm the author.
[0]https://github.com/unkn0wn-root/resterm
You can’t have a unicorn valuation built on a simple offline client.
This is basically standard SV play book - first step is „fix“ it to have a cloud component/account. Then store some of the data in cloud only - say for example for team sharing or a „workspace“. Then make some features paid only. Then ramp up the enshitification till metrics show it’s saturated and users will bear no more
[1] https://yaak.app/
I didn't need another reason not to use Postman, but I'll take it. Shame on them for moving in shadows.
Anyway, I'm sure somewhere in some terms and conditions I agreed to, it says "we own the glass".
Circumvention of the trust model of the agent harnesses (the hash), prompt injection and silently registering an outbound remote MCP endpoint
Older versions are fine, though.
https://github.com/Kong/insomnia/discussions/6590
Bruno [1] is the current better alternative to Insomnia.
[1] https://www.usebruno.com/
I have a sense that a dumb Web API testing UI application is really not that hard to get AI to implement. These apps all look about the same for a reason. So maybe there will be a prevailing FOSS option that isn't trying to monetize.
FYI the license terms only apply to the prebuilt binaries. You can build and run from source for any use (including commercial), for free.
Happy to answer any questions as well.
I'm for the most part just jaded about it all because I've seen Postman and Insomnia slowly boil the frog with increased pushiness towards logins and monetization. If a project is trying to make money at all, I feel it will inevitably grapple with the same issues. Especially for the kind of product where we can have five different options for in the comments section already. But I hope you do succeed in maintaining an honourable revenue stream.
long-time engineer here, begging others to reconsider this attitude. Engineering work has value, engineers deserve to be paid for their time, and the community will have better software if engineers are able to get paid for their time.
Is there another profession where charging a fair price for a product is looked on with derision the way it is with software?
I really like that you can paste a curl in to setup a new request.
This is definitely the type of app that some AI agent could cobble together pretty quickly, so I don't know why Postman etc thinks they have enough of a moat to keep enshittifying their product.
What can go wrong by uploading your things, if you had any passwords or users there that "pinky promise" will not be uploaded, then running AI and maybe oops have a leak and now someone is hitting your production servers?
Postman also removed the ability to export individual requests and grouped them into "Share" which means now you have to get a subscription to get your data out, and most export options are gone completely, except for collections itself - which they ask you questions why before you can get your data out. Fuck Postman. I moved to Insomnia as of last night (open source).
There's also one available as a major mode for Emacs: https://melpa.org/#/restclient
If the command line is acceptable but you want more of a focus on usability than curl has, there's httpie (https://httpie.io/) and its clone xh (https://github.com/ducaale/xh).
It'd be a personal dev app where only features we need are implemented, including features only we (and no one else) need.
I dropped them when they first started forcing you to have an account and from them only used it when work made it mandatory, in some cases we even dropped them for other alternatives (http with source in git, scripts..etc)
Personally I use a wrapper cli I built myself over curl which automatically gets api docs of my product and easily get the token so its great for agents
It is still fully native (real native, not fake web-driver native).
"No new versions of PAW are currently planned for release. No additional bug fixes, security updates, or feature enhancements are planned."
I'm currently using Bruno (https://www.usebruno.com/)
Works well enough without getting in my way.
At this point I do not see why the entirety of their software could not be a one shot prompt for the price of deer nuts.
Which, in case others didn't know, are under a buck.
Most of the YC companies are loan backed with investors pushing milestones for the next investment chunk or round.