It's wild to me that Jetbrains has been making so many top-tier IDEs, languages, runtimes and other developer products for 25 years now and is valued at maybe $5B, meanwhile we have months-old "pre-revenue" startups releasing AI coding wrappers and raising money or being bought out for twice that.
- Generating revenue from customers is about more than just creating a great product. You also have to reach lots of customers and convince them of your value. Many naive idealists think only product matters (or should matter), and neglect distribution. But most people eventually come to understand that both are necessary, and that this is practically a law of physics, not something to moralize about. (FWIW JetBrains is quite good generating revenue, and I'm fairly certain their revenue dwarfs that of Cursor and Windsurf.)
- Whoever is paying you is your customer, no matter what alternative word we use for it. If you're an employee, your customer is your "employer." If you're being acquired, your customer is your "acquirer."
- In most cases, acquirers are playing the role of investor. Investors value returns. If you want to provide value for an acquirer, then, you need to convince them of the future value of your business should it be acquired. That's usually best done through growth trajectories.
- It's perfectly valid to continue generating revenue year after year without being acquired for eye-watering sums. It's a waste of your emotional energy to become jealous or indignant when others get acquired or succeed with less work. Good for them, just keep doing you. That also goes for the rest of us in the peanut gallery. We don't need to attack recent successes to defend the honor of our favorite incumbents.
When analyzing the value of a young, pre revenue company, one of the things you want to look it is how established comparable companies are valued. For AI coding assistants, the field is too young to do that directly. However, they are competing in the space of "developer productivity tool for writing code". That space is an established market that is currently dominated by IDEs.
Seeing young companies which are pre-revenue, which are competing for an unproven yet crowded sub market (AI coding assistant) out value an established incumbent in the larger space does not compute.
Add to this the fact that there is very little moat for AI coding assistants. Assuming the market as a whole proves itself, there is a very good chance that the winners will be the established incumbent IDEs who can add AI assistance as a feature in their established products.
All of that is to say, current AI valuations in this space look a lot like a bubble.
This analysis framework you're providing would've missed YouTube (pre-revenue; no incumbent successes; crowded with competitors like Google Video, Metacafe, Vimeo, etc). It would've missed Instagram (pre-revenue; no massive photo-focused incumbents; tons of competing photo sharing apps in the App Store at the time; no moat against a big social app adding filters). It probably. would've missed WhatsApp. And many others.
Which suggests that your framework is lacking.
Here's where:
1. You're neglecting to look at the differences between the fast-rising stars and the comparable incumbents, and instead you're assuming that the incumbents automatically represent a ceiling. In this particular case, JetBrains obviously isn't the most ambitious company on the planet, and isn't focused on hyper growth. There are plenty of avenues for AI IDEs to grow and expand their revenue that have yet to be explored.
2. You're overestimating the importance of concrete moats. Google had no concrete moat either. Just because people can switch easily doesn't mean they necessarily will.
3. These companies aren't pre-revenue. I believe JetBrains is making something like $400-$500 million dollars a year, after 25 years. Cursor is at half of that in just 2 years. Windsurf is also doing big numbers.
4. Related to #3, you're underestimating growth trajectories.
5. You're leaving out the context. Companies that can afford to make $3B acquisitions (a) have tremendous war chests, and (b) have extremely ambitious goals. They're not looking to build the next JetBrains, they're looking to join the pantheon of $1T companies. Achieving massive 10x or 100x or 1000x growth as an investor/owner requires making asymmetrical bets -- bets where if you lose you're still okay, but if you win, you win big.
Reading a bit between the lines, it seems like the buyers either 1) think that ai assisted coding will get good enough that a lot more people will be doing it - that in the future companies in other fields will spend on it for their employees much the way they are paying for general ai assistants now. Or 2) more likely, they think they will get good enough to completely replace programmers, and the current coding assistant's role is mainly to gather information from developers to eventually replace them completely, by selling a spinoff product at a much higher price. They think they need spyware, and coding assistants are the best version available.
I believe the key thing you are missing here is that there is probably an expectation that AI based coding tools will be more mass market, rather than traditional developer market.
ie it's a mistake to estimate the potential upside by looking at the size of the current developer market.
As you move you're coding tools towards a less technical customer base - there are two synergistic effects - your potential customer base is much much larger, and they are simultaneously less technically competent on average ( and so less likely to build their own tools if you charge too much ).
Whether those assumptions are true - time will tell - but I definitely see more people thinking software development is now accessible to them.
- Whoever is paying you is your customer, no matter what alternative word we use for it. If you're an employee, your customer is your "employer." If you're being acquired, your customer is your "acquirer."
That is none-sens.
If that were true, all profits would go to the customers. No, you are hired because the company and its owners can extract the profits from your work. Laying you off could mean nothing to customers, but even more profits to them.
Almost everyone here is providing business value in service of these rules of the universe. Those who aren't in cost centers probably need to reflect on this reality more.
It's not a waste of emotional energy to consider how to convert a company from the $5-billion slow and steady type to the $10-billion instant acquisition type. Indeed the premise of our economic system is that maximum value is created when people continually strive to maximize the values of their companies.
Yes, and what an unstable and cannibalistic system that is! That's why some of us prefer not to sell/go public, instead opting for stable albeit less income. As another comment said, I don't need a third house or car collection. My one house is plenty, and I'm already in a position most of the country can only dream about.
It doesn't matter whether you like the system or not. It is the system, and you will be punished for not following it. In this case, by an opportunity cost of 5 billion dollars and several years. If you think you can change the system, try it and see what happens.
I'm good, thanks. the 5 billion can just stay with whoever wants them. It is this madness of "opportunity cost" that is ruining us, the idea that this is a game and optimal "play" for a single piece is desirable.
"If you show revenue, people will ask how much and it will never be enough. The company that was the 100x or 1000xer becomes the 2x dog. But if you have no revenue, you can say you're pre-revenue; you're a potential pure play. It's not about how much you earn, it's about how much you're worth. And who's worth the most? Companies that lose money."
It's a quote from a character from the TV show Silicon Valley. Russ Hanneman is a SV billionaire who made a lot of money putting "radio on the internet". He's intentionally a wild card character acting as if he has enormous business insight while often being at least partially if not fully wrong in his ideas. It's a parody of SV business ventures not having any real plan to make the business profitable, they're just wanting to show crazy growth in metrics like user counts to justify extreme valuations to sell the company and walk away before the music stops. "I don't want to make a little bit of money every day, I want to make a fuck ton of money all at once."
The next couple of lines:
"Pintrest, Snapchat, no revenue. Amazon has lost money for every fucking quarter for the last 20 fucking years and that Jeff Bezos is the king"
The wildest part is that JetBrains has their own AI coding tools and they seem to be very good.
They also have a pre-built customer base to sell them to.
I do think they have a perception issue with devs whose perspective on their products was crystallized back in the 2010s when they were using some old company laptop with 8GB of RAM when they could feel too heavy. With a modern laptop I just don’t care at all if my IDE takes up a few gigs of RAM.
JetBrains also ranks well on things like low-latency input, which surprises a lot of people. They do seem to care about developer experience.
I have truly loved using Jetbrains IDEs for Java, Go, Typescript and have subscribed to the full toolkit at personal expense for many years (it was the grandfathered pretty cheap price that ends this year). But when trying to use it for Python, the code intelligence was so poor I finally gave VSCode a try. It was much better at Python analysis at the time though I assume Pycharm has improved since then. But I'm not going back - Jetbrains still tends to have better refactoring, good features like running individual Go table tests, etc. But it's a killer feature for me in a team to be able to share a workspace or settings JSON file in a repo to get all developers on the same page. .idea has never worked well and I suspect most people assume it won't work.
If Jetbrains reimagines their settings persistence story, I'd be happy to give that a try since otherwise it's quite great still!
I am using their free product, IntelliJ CE (community edition). I simply couldn’t get used to VS code and its AI derivatives.
Possibly an interesting data point is that my company pays for every engineers’ Cursor usage, can’t imagine how much it could cost, but they don’t have any encouraged integration with JetBrains… so while JetBrains products are good, I’m wondering if Cursor simply has a better sales team and hype pushing them to higher valuations
They have such a nice product! But every release comes with almost the same number of bugfixes and new bugs. I wish it was a more stable product and they were more cautious about adding new features. Every new feature comes with the risk of adding more bugs.
I feel that. I have very mixed feelings about updates of products, and Jetbrains is no exception. It's a mix of "maybe this and that has been improved" and "they probably added a bunch of things I really don't want and have to fight for a day to get rid of".
Granted, they're not the worst offenders. When I read that Jira has been updated, I need to work up the courage to look at it because I expect it to just be worse on every level.
I don't agree with your assessment, but I want to encourage you to raise bug tickets on YouTrack (their answer to Jira). In the last 10 years, I have had more than 100 (no joke!) of my bug tickets fixed by JetBrains. If you write a good description with a min-repro, they will fix it.
> I wish it was a more stable product
Can you give a specific example of a bug that you encountered?
I can't get how poetry support problems were introduced. They have made new patch release just before Christmas holidays which introduced major problem (IDE periodically updates package lock file). It took them 3 months to fix it. After a few weeks they reintroduced the bug and said that it is feature and added registry option to revert to old functionality. I still do not understand what is the reason for updating package lock file.
You have the option to not update in place. Download the new version and keep the old one too, then test the new version, if they broke your workflow go back to the old version until the bugs are fixed.
It doesn't usually take a too much code - usually around 20,000 lines - where I personally am not able to think my way through debugging complex failures. Tombstoning code only gets you so far (and is a useful skill). Being able to step through code and jump around symbol references is a huge help for my limited powers.
If you have fully memorized all of Linux and GLibC and all your supporting libraries then yeah, I guess you don't need powerful tools.
> Editors like VS Code even, while have some conveniences built in, update every five minutes it seems.
Since I dont need updates every 5 minutes, I disable the update checker if on Windows, and just don't update my software stack all of the time as it inconveniences me.
> The wildest part is that JetBrains has their own AI coding tools and they seem to be very good.
Which one?
Last time I checked JetBrains' AI tool and it was laughably bad compared to Copilot. My bar was quite low already as I hadn't even used Cursor by the time.
Edit: What I tried is "JetBrains AI Assistant". I haven't tried Junie yet.
I've been using IntelliJ IDEA and similar products for almost 10 years, and I'm not impressed.
Java/Kotlin is their main thing, and yet neither Maven nor Gradle builds are stable. If your build fails or there are some unresolved dependencies, you restart IDE in hope it works...
AI coding tool trial failed for me -- IDE told me it's not activated even after I activated it on billing portal. And doc implied it might take some time. WTF. Does it take some batch processing?..
People who were able to get AI coding tools working said it's way behind Cursor (although improving, apparently).
There's something severely wrong with your setup if you can't get stable maven or gradle builds, and your AI problem... maybe it was really early right after release? Either way, contact their support.
And if "If your build fails or there are some unresolved dependencies" you check your dependencies and config.
I'm tired of people complaining and not trying to understand how their systems (or an IDE for that matter) work.
Because JetBrains products DO have issues, but rest assured, the things you are complaining about are on the main path of basic features they take care of the most.
Source: at first reluctant but now happy IntelliJ user, after thinking for a long time that Eclipse/Netbeans would be better. I was wrong.
> There's something severely wrong with your setup if you can't get stable maven or gradle builds
To be fair with OP, I had a similar experience in the previous company. Sometimes after bumping a dependency or something, even if you asked IntelliJ to "Reload All Gradle Projects" something wouldn't work and I needed to restart the IDE. Not saying it was common, but it did happen. This was a Kotlin codebase.
Now working in a company using Scala, I had a few cases where IntelliJ failed to resolve a dependency correctly so it was saying that the class I just created didn't exist (this was in shared library and two different dependencies pulled it, but we did pin the dependency directly and sbt was building it correctly, it was only IntelliJ that was becoming confused). Cleaning up the build sometimes worked, sometimes didn't, but even when it worked it would go back after a while. Eventually we updated all dependencies in all projects and it now always works, but it was painful for a while.
Look, I understand that IntelliJ is better than Eclipse. Eclipse is just bad. (E.g. if you open existing workspace with a newer version of Eclipse it crashes - a problem they haven't fixed for 20 years!!!)
But I can tell you IntelliJ has way more issues than e.g. Visual Studio I used 20+ years ago.
Counterpoint from me: I've been using Jetbrains tools for over 10 years as well. Mostly Webstorm and Rider and it's all working well. Sometimes there are bugs, yes, but I had plenty those in VSCode and Visual Studio as well.
Aside from their initial AI plugin rollout fiasco it has been smooth sailing for me.
Integrate it over time and you'll get a different result: JetBrains will still be with us 20 years from now, while these "AI" startups will go the way of NFTs and blockchain long before that.
I believe that's why the comment author put "AI" in quotation marks. There's a massive amount of fraud around "AI" right now, as there always is in the startup scene with a hot new technology.
there is still a distinction as the field itself is not a hoax, which I can't say the same about NFT.
Unless you prefer to think of the AI field as representable by a bunch of Indians actually behind software, as the sibling (insincerely and again lazily) reduces it to.
Engineers that want to run startups need to stop denouncing fields they don't like as "fraud". Some people made more money than you. Learn from them and it could be you next time.
Where am I reducing an entire field to anything? If one truthfully says there's a bunch of companies doing this or that, that associate AI with fraud, that's not the same as reducing an entire field to fraud.
Nobody in Russia who valued their business chose to incorporate it—or at least its parent company—in Russia. After Yukos, the writing was on the wall. (And now, of course, everybody who values their business takes pains to point out their chosen country of incorporation as though it wasn’t originally just a saner jurisdiction. As you’ve shown, everybody does it because it works.) JetBrains specifically did employ a fair number of both Russians and Ukrainians, though, and unlike Yandex or mobile carriers, was not prominent enough to be forced to split immediately after 2014, so it was inevitable they would have to move out of Russia (and they did so pretty gracefully).
There are no native Czech speakers in their management. It's a similar story as the Yandex guys going off to Nebius Group in the Netherlands after the invasion.
Sure, but they're not a Russian legal entity and as far as I can see never have been (unlike Yandex). So, to the point that their being a Russian entity might be a problem for investors, it's moot, as they are not one.
Presumably, if you wanted to stick the boot in on Russia, a decent way to do that is to encourage smart, economically productive Russians to incorporate outside Russia and attract as many young capable Russians to come with them as possible.
It's not us lumping AI together with fraud. It's the companies employing, for example, Actually Indians to do things they claim are done by machine. Or the ones marketing ChatGPT with various agentic hookups as a replacement for developers. Or...
The hundreds of AI startups desperately trying to convince management that they can replace entire teams of skilled humans with AI are, by definition, a scam. Maybe a different type of scam than crypto rugpullers, but still a scam.
No, that's fraud. A scam can be legal. Timeshares, "Winter Wonderlands", many extended warranties, planned obsolescence, weaselly technically-correct advertising (overpromising to investors is sort of this) pretty much everything hidden in ToC screeds, some MLM schemes, basically all cryptocurrency (and a lot of normal finance to be fair), etc etc.
Many graduate into criminality, but it's not required.
Most "AI" startups aren't building coding tools, and the utility of this tech goes down dramatically in industries that are less legible on the open internet than software is.
I didn't say I blindly trust generated code, it needs to be read through and covered by tests, but tasks like generate boilerplate code for some lib I never used before and which would take time to locate and read through documentation now are 10x faster.
There is huge utility in LLM outside generating complete working code.
What kills me more is originally when they released Kotlin, and I saw Kotlin Native, I assumed they would go all in on Kotlin Native, and allow you to produce fully native JRE based apps (JRE based in the sense that it can take full advantage of those libraries the JRE provides, or any plain old Java Object) and produce native and highly performant binaries.
It seems .NET already has a really mature AOT, I'm really hoping .NETs AOT reaches the point where all .NET code can be AOT'd someday.
I feel like Kotlin could do so much more, but its stuck in standstill. There's even some language features that are still missing such as Inline Classes, Pattern Matching, and even Reflection, all things that Java supports directly.
Programming languages are high-cost, low-revenue beasts. When Kotlin was first released, JB was hoping it would replace Java in the enterprise. And as the sole providers of Kotlin tools, they'd have a greatly expanded market.
Kotlin did not replace Java, except on Android. So JB now has a beast they have to feed without an enterprise revenue stream. They do get secondary benefits: they use it internally, etc.
We examined Kotlin in detail for a CLI app and based on conversations with Kotlin developers concluded that it was not sufficient of a Java replacement for us to evaluate further. For those in a similar situation, the greatly increased cadence of Java releases has probably permanently foreclosed Kotlin-qua-enterprise language.
The irony is that there are a lot of "XYZ Native" solution, such as Flutter, Kotlin Multiplatform, Xaramin, etc. But somehow only React Native, something based on a web framework, seems to reach the critical mass.
I don't really see any cross platform framework hitting "critical mass". React Native is very accessible to web developers. Are you sure that is not just information bias from your own situation? For me, most cross platform solutions are a race down to the lowest common denominator (and this is from someone who has used Xamarin Forms and MAUI in production apps.) Sure, you get to use your favourite tech stack (for us it was .Net) and you get familiar tools and metaphors, but the overall results are very uncanny valley unless you put a lot of effort in to styling for each platform.
What did you mean by "inline classes" here? My only understanding of that term applies to something like what Java is still developing for Valhalla but has not delivered yet, whereas Kotlin is arguably a little farther along with its own feature with that exact name.
Would it be the classes Java has always had that sit inside a parent class and can access the instance fields/properties of the parent? These are very Java, and hard to recreate in a language that doesn't support that type of embedding. C# really struggles with it. Using the Android API via Xamarin was always very much harder to do anything with because of this symbiotic relationship. I have done a little Kotlin, but I don't remember if it support that or not.
There are fundamental restrictions where NativeAOT will never work or be desirable. For example, runtime-compiled regex patterns or any other feature which relies on emitting IL at runtime and creating new members or even assembles and then loading them. Similar applies to compiled expression trees - they are supported in NativeAOT but in the form of falling back onto interpreting them, which has worse performance. Or unbound reflection with patterns that cannot be statically proven and/or analyzed.
Reflection analysis can (and will) be improved but there are hard constraints - a correctly working expression like 'someAssembly.GetType(Console.ReadLine())' by definition would have to root (and force compilation for) every type in the assembly, which is highly undesirable or even sometimes unfeasible for AOT compilation. And there is a lot of code which does exactly this.
The main challenge are packages and frameworks. ASP.NET Core is largely compatible (via minimal API) and so is AvaloniaUI, EF Core has some compatibility assurances and DapperAOT is tailor-made as the name implies, serialization is also a solved problem although you may need to use a different API.
At the end of the day, NativeAOT is not something "to be fully migrated to" because it has fundamental restrictions (some of which also affect other languages like Rust or Go) and having JIT around is a feature for patterns which specifically exploit it but is also a performance optimization (DynamicPGO, better instruction selection especially around SIMD paths, turning static readonly's into JIT constants and apply subsequent optimizations on top of that, this is what makes C# port of Mimalloc so good as it elides dead code with assertions impossible to remove dynamically in C/C++). NativeAOT has its own optimizations, and it will continue to diverge with JIT (e.g. there's a toggle in .NET compiler to repeat some optimization phases, usually it's too expensive for JIT but for AOT it's a good fit, AFAIK there is work to productize this).
The wide perception that JIT-compiled code has to be slower stems from other sources of performance overhead that are typical to languages which happen to use JIT (many of which have "weaker" compilers too), not from the JIT compilation itself. There are technicalities like certain calls have to be indirect in order to support patching the callee address, or inter-procedural analysis which is trivial to prove under AOT may not be so under JIT where new callers/callees may be constructed dynamically or a reJIT invoked which would invalidate the analysis results. JIT also costs additional memory. But it's not a source of worse performance.
This is the norm. The gaming industry glaringly works that way since it came into existence. There are a lot of privately owned companies creating awesome stuff that stay awesome until/if they decide to IPO or sellout to a public company. Public company owners mostly have terrible incentives and time preference which makes everything turn to shit.
I gave them a year of subscriptions before cancelling recently, the devcontainer implementation in their Ultimate versions is laughably bad, bugs upon bugs and tickets where their support staff just bounces it up with "still no fix" messages and customers are finding workarounds, i.e. downgrading docker installs.
Remote SSH is terrible too, handles network latency spikes by repeating keystrokes. I remember spending an evening trying to fix something in the
integrated shell and giving up, but sadly forgot what. I like what they do with Go though. Anyway, back to nvim here, not for me.
+1, the Remote SSH is horrible. Takes forever to connect and is extremely laggy once you have connected. Feels like they’re practically streaming video of the UI back to you instead of VSCode’s Remote SSH which feels indistinguishable from running locally.
Yeah I'm a big fan of the JetBrains IDEs but I tried it out a few days ago and couldn't even get it to stay connected. Gave up after reading a few recent forum posts about how much trouble people are still having with the feature (which does have a big purple "Beta" label in the IDE at least).
Jetbrains has "AI Assistant" and "Junie" (the latter is much like "Claude Code"). Junie is a great AI coding assistant, seems to be on par with Claude Code (and can use Anthropic models e.g. Claude Sonnet, or other models).
Also, Claude Desktop can be configured to serve Jetbrains MCP Server, which will let Claude Desktop (or any other coding AI/LLM) connect and control Jetbrains IDEs, including changing project configuration, listing / finding files, editing files, looking at VCS diffs.
So I believe Jetbrains is addressing the AI coding assistant market, they're not making as much noise and perhaps they should be ... feature-wise I think Jetbrains IDEs + AI integrations will be as good, in the long term, as other systems. At least I hope so, because I can't let go of PyCharm, Webstorm, IntelliJ IDEA, Goland, et al
> they're not making as much noise and perhaps they should be
Jetbrains isn't a silicon valley startup and isn't raising money, so no VC is going to make a 10x return by hyping them. That sadly usually means that you are shut out from the conversation no matter how good your product is.
They are a big company that has taken a lot of investment over the years. They make good stuff and their audience (a lot of them developers) LOVE the product - willing to pay their own money for them. Rather than complain that they're shut out of the system we all apparently don't like, shouldn't we celebrate that they're making it with an alternative approach?
I've had a Jetbrains license for years. I barely use their stuff now. Cursor is much less capable IDE but better coding environment. At first I'd use them both on the same thing but now it's all Cursor.
Exact same here. I love PHPStorm for Laravel dev, but I’ve barely used it in months. Meanwhile I’m paying $100+ per month to Cursor for subscription + premium model usage. And honestly I kinda hate Cursor as an editor / IDE. I really hope Junie is good enough to switch back, but I won’t give up integrated AI agentic coding in my editor if it’s not.
That's pretty much the comparison between Tesla and old-time car manufacturers. Most people who are trading Tesla stock don't even look at other car stocks.
Because the products being good isn't what investors look for. Investors look for good business. And, assuming they can actually make it work, putting together AI coding apps that get software "engineers" to outsource enough of their thinking to a machine that they can then jack the price through the ceiling on and make bank via enterprise billing is incredibly good business.
This does seem very unfair, but don't believe any of the numbers being used to value these OpenAI acquisitions. If it's not cash, it's just a lottery ticket.
It's the difference between bootstrapped companies and VC backed or public ones. Bootstrapped companies don't have any growth expectations whereas the funded ones do and are priced based entirely on speculation.
The only good investment in crypto has been to buy and hold BTC. So while it has worked well as an asset (like gold or paintings or baseball cards), the crypto industry itself – thousands of startups in the space, all the altcoins, blockchains, smart contracts, web3, NFTs, ICOs, DeFi, "stable" coins – have all either fizzled out or just been outright fraud.
If price of gold was according to its "actual value", it wouldn't be $3300/oz right now. Nor would anyone hold it if they expected its price to ever get down to such "actual value". Bitcoin really isn't any different in that regard - it has value because people are willing to trade it. Although it does have practical uses, as well - black markets, for example.
The actual value of bitcoin is that you can use it to buy illegal counter-cultural things like drugs, fake IDs, etc. Except you can't, because more anonymous cryptocurrencies were invented and the markets migrated to those. Now Bitcoin is mostly used for more "serious", "wall-streety" crimes, like bypassing international sanctions. Either one can hold up the value of an investment asset.
I don't know how anyone can still say Crypto was "hype". Sure the vast majority of it is but USDC alone market cap is north of 60 Billion, there is nothing speculative about USDC. You can't make money by someone else losing etc etc. Tether stablecoin is ~150 Billion. Just those two, means the value of Crypto minus rug pulls, scams or anything else is over 200 Billion - more than the value of OpenAI
I for a while was completely into crypto and really hoped it would replace a few mega corporations and enable easy convenient transactions… but that never happened. I can’t buy groceries or pay rent with any crypto. Infact In fact I regret spending the crypto I did have on the very few things or places I could spend it. Crypto(at least imo) went from something new and exciting and limitless to now being only something you buy and then hope the value goes up. It’s useless as anything else. It’s now all snake oil salespeople trying to pump up some shitty crypto that has 0 new features and was created to pump and dump.
Aw shit, I guess I should stop doing actual products and just go where the fad is. That's what the "smart money" does, I guess. I mean, isn't it smart to pursue maximal profit? Max payoff - you can use it to fund other things!!!
You're being sarcastic but yes. If your intention is to take large sums of cash from investors and run off with a fat check then that is exactly what you should do.
No I'm actually not being sarcastic. I know it might sound like that - it's hard to grok tone from text and I'm not particularly concerned. But I'm seriously considering it, I mean -- the goal of most startups is to solve (as PG puts it) "the money problem". So what am I doing building products, when I could be chasing these fads and possibly banking huge? I mean, that's smart, right?
I'm not being sarcastic at all. I feel I need to reiterate that because you got it wrong.
If enough of the market participants act stupid, or at least in predictably irrational ways, it may be profitable to capitalize on that behavior.
If you look at previous fads, all the way to be dotcom boom of late 1990s, this very approach seemed to work well for a number of buzzword-compliant pre-revenue "businesses", and their founders / owners. There was a crash after that, but the wisest were able to shield some of the money from it.
I think there’s that side to it, it may be correct to swoop in and profit, if you can outsmart them.
But then there’s another side where the fad has real value and the people investing have money because they’re smart and they recognize the value. And you have to be to make value in the fad.
But on some level, it’s also stupid to pursue the fad because it’s so hyper competitive and effect of luck is going to be magnified.
Some people might say there is more to life than making money but if you just want to hopefully get some naive investors to give you a large pile of money for your vaporware then go for it, it's where the money is.
Thanks, I appreciate your answer. Not sure it’s me tho. I guess another side I’d there’s probably real value in the fad, too. On some level, if you can figure that out.
All these people with money didn’t all get it by being stupid or fddy or vaporware so they’re responding to signals of value and their responses are reliable indicators in general.
There is more luck involved with getting rich than you might think. Mark Cuban once said if he had to start over he would have no problem becoming a millionaire but becoming a billionaire is completely dependent on luck.
People entrusted with investing billions of dollars are very much not rational, as we can observe every day.
Additionally, network effects and experience input and VC concentration in Silicon Valley are very real and very rational. For VCs, why go anywhere else if the best of the world are flocking towards you already?
This might change in the near future, but I doubt it.
Not to mention that JetBrains is implementing AI agents into their current products -- if I'm already a user of IntelliJ, PyCharm or CLion, etc., why would I switch to Cursor or Windsurf?
People love to crap on VC but it's a great transfer of wealth from the rich that have so much money they can give it to other people to gamble for them by paying people to drop out of school to build things that will never make a return on investment. \s
Not a direct fork but VS Code was built on the foundation of Atom, and heavily inspired by it. People generally consider VS Code to be the successor of Atom, especially after Microsoft bought out the creator of Atom/Electron (Github) and shut down the project.
VS Code is built on the foundation of Monaco. And when you run it in a browser it has nothing common with Atom. And on desktop the only thing from Atom is the runtime Atom shell now called Electron and it is essentially Chrome with some glue for integrating with OS.
Microsoft built several key tools on top of Electron (VSCode being the relevant one here) and became very interested in maintaining control of it... and so bought GitHub when it was up for sale.
> Atom has not had significant feature development for the past several years, though we’ve conducted maintenance and security updates during this period to ensure we’re being good stewards of the project and product. As new cloud-based tools have emerged and evolved over the years, Atom community involvement has declined significantly. As a result, we’ve decided to sunset Atom so we can focus on enhancing the developer experience in the cloud with GitHub Codespaces.
> This is a tough goodbye. It’s worth reflecting that Atom has served as the foundation for the Electron framework, which paved the way for the creation of thousands of apps, including Microsoft Visual Studio Code, Slack, and our very own GitHub Desktop. However, reliability, security, and performance are core to GitHub, and in order to best serve the developer community, we are archiving Atom to prioritize technologies that enable the future of software development.
JetBrains AI now does essentially the same thing as Windsurf and Cursor. My prediction: JeBrains IDEs will be around in 5 years, Windsurf and Cursor won't.
As someone with decades of software engineering under my belt, my opinion is they are the best IDEs out there by a comfortable margin. It's possible the issue you're encountering when accidentally opening them is due to index updates which would occur disproportionately often on startup for people who otherwise never use them. Other than that, performance is more than satisfactory 90+% of the time in my experience.
That's what I would have said about Eclipse about 15 years ago.
I rarely have performance issues with Jetbrains IDEs, and even rarer still that they bog down the whole machine. Get a properly specced computer (32+GB RAM, 10+ core chip) if you don't already have one. Also try adjusting memory settings - https://www.jetbrains.com/help/idea/increasing-memory-heap.h...
That's just it. In my company there are basically three sets of developers. The very experienced ones prefer vim/emacs. The experienced ones tend to like these more established IDEs, and the younger developers absolutely hate these IDEs and find them to be slow and bloated.
I can't really comment on whether they are good or not since from my perspective I see people who are productive and unproductive using both of these tools, so to me it just looks like mostly a matter of preference. But newer developers don't seem to like these established IDEs and see them as you said: big, slow and laggy.
It seems like JetBrains may be giving up on Fleet. Honestly though, I think their established IDEs are good enough to be worth a few bugs and performance issues.
Intellij is an IDE where VS Code is a bit mroe then a text editor.
When you open a project for the first time it takes time because it is indexed , maybe it will index your dependencies if you did not bother to disable that.
IWhen i open a file /project from a collegue that uses VS Code is filled with errors and warnings because their VS code text editor is not actually understanding the code they are editing.
People that say this are completely out of touch. You know there are VSCode extensions for Java (made by Oracle and Redhat themselves), right? Once you install that you get Intellisense, debugging, etc. Same with C#, Python, etc.
I know that, once but once you install that you will not be able to praise VS Code performance anymore, you just get an inferior Intellij IDE in all dimensions.
I'm super happy JetBrains has been opening up all their editors to offer free access for non-commercial use. I've never made any money off of my occasional side projects, so I could never justify the cost to pay for a license when I may go an entire year without using. But I really love their editors - their key mappings are an extension of me at this point, it's so smart about figuring out how the code works and letting you find usages or refactor without thinking, and their git UI is basically the only way I find git tolerable.
I may or may not have been abusing the fact that my university let me keep my email address as an alumni to squeeze more years out of their free access for students, though that seemed to stop working for me at some point a year or 2 ago. But I'll happily take this instead!
They have this terrific plan where if you buy yearly subscription you will get the major version free for life even after you stop paying after one year.
I was pleasantly surprised when I discovered it. Been using it since then for like 3-4 years now.
It's CLion, Rider, RustRover, and WebStorm, IntelliJ is not on the list.
So far they haven't muddied the waters for any versions that already had free Community Editions (IntelliJ and PyCharm). The Community Editions are more limited, but don't restrict commercial use.
> The Community Editions are more limited, but don't restrict commercial use.
I've always wondered about this. I have the All-Products Pack subscription, don't get me wrong, but I used to have the Educational licence when I was in university. What was there to stop me from using it for commercial purposes? I get that the licence restrictions are likely more targeted towards medium to large businesses than little ol' me, but to what extent is it just an honour system? Just don't commit your .idea/ folder and basically no one would have any the wiser?
> I'm super happy JetBrains has been opening up all their editors to offer free access for non-commercial use.
All? That would be news to me! From the 10 IDEs (not counting ReSharper, which iss a plugin vor Visual Studio) listed on https://www.jetbrains.com/ides/#choose-your-ide, only CLion, Rider, RustRover and WebStorm are free for non-commercial use. Plus, each of the products has its own free or discounted licenses for certain users (e.g. students).
"Has been opening" is the present perfect continuous tense [0], which describes something that started in the past and is still ongoing. In TFA JetBrains says explicitly that this is a process that they intend to continue assuming it goes well.
I have been using PyCharm, IDEA and Android Studio for free for a while now. I'm not a student or any special category of user. I think you only get "core" features for free but they sure still among the most featureful IDEs that I have used.
I'm not sure why the staggered rollout, maybe there's strategic reasons certain ones will never have a free non-commercial license. But so far they've been consistently opening them up one-by-one.
It'll be interesting to see what they do for PyCharm and IntelliJ, which already have free Community Editions. Long term, I doubt they'll want to have two types of free version that restrict usage in completely different ways, but if they kill the Community Editions, anyone using them for commercial use will have to either build from source (hopefully the end of community editions wouldn't mean the end of the open source parts), start paying, or switch to an alternative.
I'm making zero predictions about what they'll do, there's a lot of ways it could go.
I started using Intellij with the 3.0 version, I think. It just worked, even on Linux. (It was existence proof that you could build excellent UIs in Java.) Unlike Eclipse, and other forgotten IDEs that were so bad I discarded them immediately. Even early on, their refactorings were usually flawless. While I think I found one screwup, they were so good that they changed the way I coded. I could easily and reliably do refactorings that were otherwise pretty time-consuming and error-prone. I have continued using their products: mostly PyCharm now, and occasionally CLion.
Each new release improves the UI, and occasionally adds features that I find useful, and many that I don't. I suspect that I'm not alone in using a very tiny portion of the features they offer. How they can keep up with all the languages, and libraries, and frameworks is beyond me, but they seem to do it.
Their support has always been excellent. I once (v4?) complained that refactorings did not extend into configurations. E.g., if I rename a class Foo to Bar, then the runtime configuration running Foo didn't reflect the change. I reported it, and found a fix in the next release. Email with technical questions or bug reports is always handled promptly and thoughtfully.
They have always provided absolutely fantastic products for free. Yes, you gave up some features, but the free versions are really useful. I'm retired now, but continue to pay their licensing fees every year, for my hobby usage, because it's worth it, and they earn it. And the licensing is not onerous to use. What I really like is that you don't have to be on the internet to use their products, just for the license check. I wish all licensed products did that.
And beyond all this: They haven't sold out. They are one of the very, very few for-profit tech companies that have maintained a stellar level of product breadth, depth, quality, and support for such a long period of time. I'm sure they could have cashed in, sold to IBM and the product would have just rotted away, (sorry, IBM, but you know it's true). I can only think of one product that is comparable in this way, and that's Postgres.
I was expecting you to end with "and don't squander the amazing track record of developer goodwill you have accumulated by forcing the stupid-ass 'New UI' on us, or telling us that we're committing code wrong, or somehow suddenly trying to go 'All AI' or whatever"
I'll be straight: for me personally there is no replacement, and I'll just patch the "unsupported" Commit Like a Sane Person plugin[1] indefinitely if I have to, but ... come on, don't strain our relationship like that
These are so interesting. From my perspective using PyCharm and RustRover: By far the best code editors, in terms of introspection and refactoring. The only ones I've used that model my projects correctly; VsCode and Sublime etc make it feel like I'm editing files, where are IMO the wrong abstraction.
I experience major performance problems. They periodically bring my 9950 CPU to a crawl, or freeze, requiring a force-kill. (RustRover more so than PyCharm, but both are guilty). Memory hogs. (Feels like they leak memory). This is consistent behavior over the years, across a range of project styles.
I put up with the performance problems because of my first point!
The interesting/amusing part to me: My experiences do not seem wholly consistent with other users: Many users seem to find these IDEs heavy, but don't experience the freezes, crashes, or memory leaks. And many (most?) people claim VsCode is fine for managing multi-file projects. I don't know what to think!
I'm using RustRover. It's pretty lame compared to the IntelliJ experience. "Find usages" does not separate test code and application code; "copy reference" gets me the file name and line number instead of the fully qualified name.
I'd probably use vscode/cursor fully if I weren't so used to the JetBrains environment.
That's insane, I use webstorm, pycharm, and Intellij, sometimes all at the same time, on a mobile 8th gen i7, it doesn't lock up. The single thread passmark score is literally half your cpu! :D
Are you experiencing these problems on your personal device or is it a corporate provided one full of scanners? IntelliJ is definitely slower on my work laptop compared to my private desktop. Might also be due to the larger project size and heavy usage of Spring, though
I have my work project loaded on my home PC, and to be fair it's a beast of a PC, but still it's insane how much power is wasted on the work macbook from the stupid MDM software running on it. It feels blazing fast on my PC, and the slowness is unbearable sometimes on the work laptop.
Oh hell yeah! I used CLion about 6 or 7 years ago at my job, and it was a pretty great product for small projects. It used to slow down really bad for a medium-sized project though and I switched to VSCode.
I've since moved on to new employers, but I'd love to check it out again.
> It’s important to note that, if you’re using a non-commercial license, you cannot opt out of the collection of anonymous usage statistics. We use this information to improve our products.
Well, it's basically true for MS-branded VSCode too. I now use VSCodium.
But I'm heavily against Microsoft. I don't like usage statistics collection, but at least this is a direct competitor to Microsoft.
I had a chance to speak to some of the JetBrains folk at CppCon a couple years back. It was really nice and reassuring.
I'll check it out for personal projects and see if it's improved since years ago. :)
> > It’s important to note that, if you’re using a non-commercial license, you cannot opt out of the collection of anonymous usage statistics. We use this information to improve our products.
> Well, it's basically true for MS-branded VSCode too. I now use VSCodium.
How's that "basically true"? That's false. You can opt out. In fact there's very good documentation around that
Personally, I think it's a shame that JetBrains get such flack for collecting telemetry in their free products when Microsoft do the same in VS Code with hardly anyone voicing the same level of criticism for it.
Assuming that the plugin is enabled for the free version, CLion is also amazing for Rust. Thanks Jetbrains!
Here's hoping this won't be abused by smaller companies that will no longer want to pay for the actual subscription. I also wonder if they are moving towards a different funding model, since the IDE space is pretty competitive with a free alternative (VSCode) out there.
Yes, it is quite nice. That being said I keep a little statistic about IDE usage at the Rust events I attend. I have observed RustRover or CLion only three times at the 48 events I've recorded.
One of these three was an event at JetBrains. To be fair, I started my notes long before RustRover existed.
Using "Java-based" as a smear is pretty silly. The IntelliJ platform is an extremely solid piece of software, clunkiness and all. It does everything pretty well, with very little need for config fiddling or third-party plugins.
I won't argue with the logic, if it works, it works. But still, I tend to react negatively to "Java-based" things. I think it comes from fiddling with Xms Xmx in the past with Elasticsearch which gave me the sense that the JVM is a beast. Now you want me to use the same beast in my editor? Neovim feels a lot lighter in this regard.
I got 'lazyvim' (a neovim distro) up and running, installed LSPs for my languages of choice (mostly java and go) and it seems like a completely broken experience to me. I don't even get any error messages telling me anything is wrong, it just doesn't work.
This is the most silly comment I’ve read in a while.
If your software requires user to debug something for you then you’re the one who should consider learning how to write software before writing software.
I use new tools all the time, almost on a weekly basis. Most of those tools have bugs, software isn't perfect. Our job is to be good at finding and fixing issues in tools, and the beauty of open source is having access to fix bugs were you find them.
It's extremely odd to consume free and open source software in any capacity and just not have cultivated the understanding and desire to improve all of the things you touch. It's one of the worst trends in programming.
Imagine a world were every tool you used including your programming language cost money, and you had to spend 5000$ a month just to be a developer, and any bug you encountered could only be solved through a support ticket. That's my idea of a terrible nightmare, and why I wouldn't touch something like IDEA with a 10 foot pole.
You know, it's funny, when I was a cs student I always wondered why some of my best cs professors didn't bother using the latest and greatest tooling. As I got older I realized, my time has actual monetary value, and spending time dealing with finicky software is often not worth it. Intellij is like ~ .50c / day, and I'm way more productive with it than I am with vim (and I say this as someone who used to mostly work with straight vim in college). Maybe you don't value your time?
People in academia are notoriously disconnected from real software development, so I wouldn't use them as an example of what to do or not do.
You can be judicious with your time without avoiding any inconvenience to accomplishing a goal. Inconvenience is how you find motivation and learn new things. For 50c a day you're robbing yourself of valuable knowledge and bettering open ecosystems that are themselves fulfilling ends. If you don't put a value on learning or contributing back then I don't see why'd you'd want to be in this profession in the first place.
You may not be able to think of a worse experience, but a lot of "newer" programmers may not even know what an LSP is. While it's true that I no longer need to rely on some of the benefits of Jetbrains, when I was getting started jetbrains paving over toolchain difficulties was invaluable.
Indeed, this is one of the various options one has.
Another option is to not use it and be vocal against telemetry, hopefully convincing others to do the same while dissuading other developers (especially in a forum like Hacker News where people that build stuff gather) from adding it on their products.
Genuine question. Why do you care if other people agree with you about telemetry? I almost always enable telemetry on anything I don't believe will serve me ads.
I wouldn't take people seriously who are vocal against telemetry in a product which has paid version with the option to turn telemetry off. Such people would be vocal against paying for the work others do. Also, if properly anonymised, telemetry isn't the devil people make it to be.
> I wouldn't take people seriously who are vocal against telemetry in a product which has paid version with the option to turn telemetry off.
You are conflating two separate things, a product does not need telemetry to function and if telemetry is needed it can be opt-in. Similarly a product does not need to be free to have telemetry nor does it need to be paid to not have telemetry, as i already wrote these two are completely separate.
> Such people would be vocal against paying for the work others do.
Again you are conflating two completely separate things: people being concerned about the privacy implications of telemetry (both directly and indirectly, see below) and people who are against about paying others for their work.
> Also, if properly anonymised, telemetry isn't the devil people make it to be.
Even if anonymized (which is something you can only guarantee for open source projects that either you or someone you trust has checked they do such anonymization properly - and also you either build yourself from the source that was checked or you used a binary from a reproducible build) having telemetry in place still creates and reinforces a precedent of it being acceptable which in turn can be used to excuse other programs doing the same but those programs actually not caring about doing proper anonymization (at best) or even outright spying on you (at worst).
Besides anonymized data can still be used in conjunction with other data to be deanonymized and the best way to protect users from this is to not collect that data in the first place.
I'm not conflating these two things. I didn't say they have to have telemetry to keep it free. All I am saying is, if you have the option to pay and decide whether you give them telemetry or not, then there is no reason being vocal about them giving you more for free. It is a company and they need to make profit. If more users with telemetry enabled gives them more data that they can use to indirectly increase their profit, I only applaud them. If they stopped allowing to turn telemetry off for the product in paid version as well, then you would have a valid point.
That said, nothing wrong with being vocal about privacy and high standards in collecting usage.
Privacy is a fundamental human right, not something you get to withhold from people until they give you money.
Yes, I could pay them to get a product that lets me disable telemetry. I'd much rather just use something else, so I don't have to fund their unethical business practices.
That is an extremely disingenuous framing of the topic. They make a product, which you can pay money for. If you prefer, they will also let you pay by sending them usage telemetry. In neither case are you being deprived of a human right or extorted in any way - it's a transaction where you receive a benefit in exchange for something you give to them.
It's nonetheless useful for people to warn others about spyware. You can find the tradeoff acceptable (or be willing/able to put the necessary isolation in place) while thanking the other commenter for the heads up.
I wasn't aware they were using telemetry in the free versions until this thread. Why I didn't just assume that a free service or product offered by a for-profit company isn't doing _something_ to extract value is beyond me.
Maybe it includes a list of fonts installed on your system, your screen resolution, etc. You don’t need much to get a “fingerprint” that is anonymous but can be correlated with those collected by other tools’ telemetry.
In theory yes? But if state actors, the ones with the sophistication to literally build a signature based on your fonts using your IDE, and then infiltrate a second application to do... whatever... if they want you that bad, they're eventually just going to get you, even if they have to just send someone to your house with a rusty wrench to retrieve all your passwords. Those guys can get the job done much more cheaply anyway.
Compared to most UI dark patterns and scummy tactics to get you to "consent" (actually tricking you into agreeing, because nobody can't be bothered to jump through like a bunch of legalese and dialogs), them just giving you that choice of straight up paying feels better.
Not really interested in their services, but at least that sort of payment would let me expect less trickery in the future.
> Critics of this consent model have called it "pay-or-okay", claiming that the monthly fee is disproportional and that users are not able to withdraw their consent to tracking as easily as it is given, which the GDPR requires. Massimiliano Gelmi, a data protection lawyer at NOYB, has stated that "The law is clear, withdrawing consent must be as easy as giving it in the first place. It is painfully obvious that paying €251,88 per year to withdraw consent is not as easy as clicking an 'Okay' button to accept the tracking."
Under this model, you'd just have to refuse service to everyone who doesn't pay (killing your platform) or let people partake in your platform with no revenue off of them (killing your platform). Neither seems reasonable from the perspective of that business? Are they just supposed to find other ways of monetizing their users or perish then?
Not that people are obligated to use IntelliJ IDEs, but it's sad that it boils down to "You can have privacy if you can afford it". But admittedly is better to have the option to use it than not being able to use it at all
Their telemetry promises not to collect private data. Yes, your code will probably used for training their models. But so it would be if you publish it on GitHub.
All data on my computer is private unless I specifically make it public. Data thieves like to make a rhetorical sleight of hand where they say they're not really collecting data about you (this happens especially with the topic of "differential privacy"), but that's just gaslighting (i.e. trying to manipulate you into thinking you simply don't understand what they are doing). e.g. I'm not willing to share noisy correlations about my preferences either. That information is private, and it is information, or they wouldn't want it.
My privacy is indeed differential. I am willing to give them the information on my coding patterns and even non-commercial code for a free license, if it is not linked to my identity. This is a fair exchange. I am not willing to do this if they will use this information to sell me ads, or sell it (unless properly anonymized) to some other company. And most certainly I won't agree if they collect any information beyond what's happen in the IDE.
Not _everything_ I do on my computer is fully private. I apply much stricter standards to things that are _really_ private. But not everything is like this.
This comment is public. It will probably be used to train yet another LLM. I am fine with that.
Sure, but as I said elsewhere, it's still important that people point out that past the headline ("it's free") is that it is also malware (it spies on you). People were free to not use BonziBuddy as well, but it was rightfully characterized at the time as spyware. If the product also functioned as a proxy for botnet traffic, you wouldn't simply say "well you're free to not use it". You'd say "beware, the 'free' version is malware". Spyware is similar.
Posting to a public online forum is of course specifically making the post public.
You don't have to use their products, though. Nor is there a shortage of free C++ IDE alternatives that don't have telemetry in them, for those to whom it is important.
“With the new non-commercial license type, you can enjoy a full-featured IDE that is identical to its paid version. The only difference is in the Code With Me feature – you get Code With Me Community with your free license.”
“We appreciate that this might not be convenient for everyone, but there is unfortunately no way to opt out of sending anonymized statistics to JetBrains under the terms of the Toolbox agreement for non-commercial use. The only way to opt out is by switching to either a paid subscription or one of the complimentary options mentioned here.”
Also, if they find that unfortunate, why did they make the product do that?
From the link: "It’s important to note that, if you’re using a non-commercial license, you cannot opt out of the collection of anonymous usage statistics."
The telemetry is used to analyzed the most used / unused features and to improve the product. For example, it's useful to understand that some specific technology gains more popularity among users and contribute more into its support.
I have no doubt that this is the primary use. It was the only way to use that data until now. Now that there is a free offering, there is a new potential use that neatly explains why it is required in the free offering and optional otherwise.
Then you're presumably committing an actual crime, or at least, their lawyers will see it that way, probably, as copyright infringement because they only gave you a license to use the software if you left the telemetry on.
As good as LSP and other tooling has gotten recently, there are some C++ projects that just need a proprietary IDE for proper code navigation and completion. A lot of times my choice is a) spend 3 hours debugging why neovim's LSP is putting squiggly lines everywhere or b) just fire up CLion.
Huge benefit of current rise of "C++ killer" languages is that they're designed with LSP support from day 1. I had a great experience with Rust and rust_analyzer: open any project in LSP-capable editor and it just works.
Are they though?
* Rust was designed before even vscode was released
* Zig, Odin, JAI are all designed by people that don’t use LSP and don’t like it
* Don’t know about Carbon, but it currently doesn’t look like it’s gonna kill anything
It seems to me that rust analyzer is something of an exception
Zig has on the roadmap the creation of a compiler REPL to power all kinds of code intelligence tools, including LSPs. This item is blocking only on finishing incremental compilation.
Sure and the incremental compiler work is genuinely amazing. I was objecting purely to the notion that this new crop of systems languages is “designed from day one” around LSP, which doesn’t seem to be quite true
Great news. It is beyond me how people are complaining about the free version not allowing to turn off telemetry. Why don't you stick to the paid version if you are bothered by the (anonymous) telemetry?
You aren't a product? Then pay for the products you use instead of asking to get it for free. If they don't give you the option to turn telemetry off in the paid product, then go ahead and complain. But pay for what you get.
? No one asked them to release their software for free. They did it of their own accord.
They don't get to subtly mine value out of their users just because they released free stuff. That's nonsense. We're not obligated to be guinea pigs in their usability experiments.
I get to do all sorts of things. For example, I get to completely firewall their software off of my network. That "take it or leave it" nonsense simply doesn't work on my computer.
Data is not a valid currency by which payments are made. Period.
You're acting like they're gonna go out of business if they can't get their little statistics. Come on now.
There is absolutely no need for me to "prove" anything to anyone. No need whatsoever. It is my computer, and I have decided. It's as simple as that.
Why? Because I don't want it to happen. That's all there is to it. Couldn't care less what their motivations are. Couldn't care less how "justified" they are. I simply do not want data of any kind to be compiled and exfiltrated to anyone for any reason whatsoever, unless I explicitly command my computer to it.
> It’s important to note that, if you’re using a non-commercial license, you cannot opt out of the collection of anonymous usage statistics. We use this information to improve our products.
Same reason responsible technologists warned less-informed users about installing BonziBuddy. It was a free product that told jokes or whatever just as it said it would. It was also spyware.
Responsible technologists should raise the alarm on spyware products because they are harmful toward their users. Malware is often given away for "free" (sometimes even sent to you without you asking!), so it doesn't really make sense to say "well that's the deal". Somehow people seem to be forgetting this over the years (I suspect because a lot more technologists make money from participating in the surveillance/malware economy these days, and it's gotten so bad that some of them have started to think malware distribution and exfiltrating (and often selling) user data is not a thoroughly black-hat activity).
If you're okay with adware or spyware or crypto miners or botnet proxies or whatever else running on your computer as a form of "payment", great. You consider that a reasonable "transaction". Other people appreciate being warned about such behavior. In any case, one shouldn't consider the product to be "free" as advertised.
Gathering user consent is what makes the difference between malware and not. If you click "Yes, upload this crash report" or "Yes, upload stats on what buttons I click", that's the program acting according to your wishes. If the program gathers and transmits that data without you asking or reviewing it and against your wishes, that's malware (i.e. malicious software that causes the computer to undermine its owner). Basically, does the computer obey the owner or not?
"It’s important to note that, if you’re using a non-commercial license, you cannot opt out of the collection of anonymous usage statistics. We use this information to improve our products. The data we collect is exclusively that of anonymous feature usages of our IDEs."
I'm aware it's common practice, but it's always good to read the fine print.
Do you think JetBrains are making money of usage statistics of their IDE?
It's much more likely they use those stats to know which parts and features of their products are used, and therefore which need improvements/love/prominence.
I have no idea what they're doing - only that now, the product contains telemetry and if I want to use the free product, I cannot opt-out. I don't consent to being tracked.
So... not that it really matters I imagine, but JetBrains doesn't do pure subscription. Rather, you buy your IDEs, and effectively have a perpetual license to the product you bought. You get free upgrades, and if you keep paying, your license gets updated along with it.
If you don't continue paying, you still have the IDE you paid for. You just don't get updates. This seems like a subtle distinction, but I think it's an important one in the world of subscription services.
The yearly subscription includes the "perpetual fallback" license GP is talking about. With monthly, you get that once you've been subscribed for 12 months.
And I recognize that I'm biased as a JetBrains fanboy, but I think it is just a stellar deal to pay $173/year for the all-you-can-eat pack where you get to keep them if you stop paying
I know folks are going to call out that the entry price is $289/year, but I still think for the number of included products that's still a deal
I just wish they offered "3rd world economy" prices for those of us not living in rich countries. I would even be totally ok with providing my National ID card to prevent fraud, just like with Hetzner.
The all products pack costs an entire (software developer) month's salary where I live.
I love jetbrains IDE for all sorts of features (their git tools like diff checker are unmatched imo). I've been wanting to use them for C++ for a while now specifically because I find that many LSP solutions are not that great. Looking forward to using the non-commercial version of clion!
Pycharm is my goto for Python; I happily use AndroidStudio when working on our Android apps.
In the past when I tried CLion, I found that its need/desire to use cmake prohibited me from really using it. We have our own build scripts, and it seemed to struggle with that. Anyone know if that CMake bias still exists?
I ended up using Nova on my Mac for C code and have been pretty happy with that.
I would really really really love it if there was an Elixir skin for Jetbrains tools.
clion still needs a cmakelists.txt that's good enough to index the project, but it doesn't have to produce a usable artifact -- you can still build with make or whatever else you want. Probably cold comfort if you use a lot of third-party libraries, but otherwise it's a fairly set-and-forget kind of thing.
I previously saw in their release notes that Makefile-based projects were out, but while digging up supporting links it seems their project formats list has gotten quite sizeable https://www.jetbrains.com/help/clion/project-models.html
It has been a while since I last used CLion. I got Nethack working with it (I did have to run it through cproto first) and at the time it auto-generated a broken cmakelists.txt that I simplified with an even more broken one that slurped up every single source file with a couple wildcards. Couldn't build that way of course, but it was enough to index all the things while the makefile did the real job. Good to know that's not a necessary step anymore.
Very cool and I am quite thankful for Jetbrains free contribution! I understand peoples concern with telemetry but I can understand that there is some kind of trade-off from a free product, my wish is that they do the same with RubyMine.
Another thing I would like to discuss is this (it's kind flimsy so bear with me), Jetbrains IDEs are good and clean. One thing I dislike with these IDEs is that they do a lot in the background that feels like is outside of my control. I can't really point my finger exactly to where this feeling is coming from but I only get that impression when I use these heavy IDEs.
Say I want my development environment / settings / workspace config to be easily replicated and configured when someone new joins the project, will their Jetbrains IDE recognize it and prompt the user to adapt? Because that is one thing I enjoy with vscode, EVERYTHING is under my control and I can see what everything does. The buttons that runs different tasks can easily be viewed on .vscode/tasks.json, all config is within the .vscode folder!
In Jetbrains IDEs, some things are local only and you have to ask the peers to screenshare which config they had. I don't want that, I want the whole workspace to be easily replicated and shared.
I know Jetbrains is working on Fleet, I've been trying it out a couple of times and I like it, I hope they take this into their consideration.
Incidentally, I recently tested Zig and I decided to try out CLion as the Zig IDE. Seems to work great with the ZigBrains plugin. It's still in development but ready for use.
I'm curious about what non-commercial means in practice.
> Common examples of non-commercial uses include learning and self-education, open-source contributions without earning commercial benefits
What if I start writing code, let's say 80% of a codebase, then for the next 3 months I switch to another editor to write the next 20%, and then commercialize the support (so open-source but with commercial benefits)? Would it be about intent, i.e. it'd be fine if I had no plan to make a business out of it at the beginning, but as soon as there's the idea of a business I should have switched?
I guess in practice this mostly targets companies with 10+ employees so it's fine not to draw the line that clearly?
It means "stay under the radar if you are cheap". This is Jetbrains, not Oracle, there isn't an army of lawyers out to get you the second you breach the agreement.
Great, but how about letting me have my paid, licensed Rust rover open on two PCs simultaneously?
I have my work PC which I leave on pretty much 24/7 and if I forget to close RustRover, and try to launch it on my desktop (for personal project, or to just work from another room) I get an error that I already have a licensed copy running and it closes RustRover. Sometimes I wake my Laptop from sleep and it does this because I still had an old window open. Really unnecessary...
> This works fine on my personal license with different usernames.
I just checked on my two machines different OS, different OS usernames.
Both copies of Rust Rover are signed in with the same account and have the same license.
I just tried launching RR on both machines and... I didn't have the issue that I've been struggling with for what seems like forever. This WAS definitely 100% an issue before though, I was trying everything to find an different editor to switch to.
The article you linked was "Updated November 5, 2024 at 11:17 AM"
So its possible they made a change since I stopped trying to do this and got in the habit of closing RR out on each machine... and didn't realize they actually did fix it.
I just bought the annual license for all IDEs did not know this was a restriction that’s a bummer. Surely they can do some pattern analysis to determine which systems belong to a single license.
For personal licenses it's a non-issue. For commercial, use the same username on both computers. If you have floating licenses, you need multiple licenses.
I used to be a CLion user. Unfortunately, it could not keep up with our large and heavily templated C++ codebase. Although CLion Nova with the new ReSharper engine improved things a lot, we found our custom clangd with emacs/vim/VSCode to still be much faster and convenient.
I know CLion also has clangd, but I believe it's their own fork. I am also not sure if you can enable all clangd features since it's not the main engine. I'd be happy to hear people's thoughts about this.
My experience has been than complaining into HN does not result in changes to software. Have you submitted a profiling dump to YouTrack?
I have a lot of sympathy for someone trying to make an IDE (or any introspection tooling) because the number of ways humans can come up with to organize or author code is unlimited. So, I'm sure they would welcome feedback on "hey, watch out, modeling templates using this mechanism has bad perf in this IDE component"
Not complaining at all! I have just heard of other people facing the same issues and assumed that this must be a well-known problem. It might also just be template heavy projects -- maybe it works great otherwise. I don't have much experience with CLion in other contexts.
It's a commercial product. Expecting people to work for free for a billion dollar company seems a little out of touch. JetBrains can contact GP and discuss rates if they value that data.
If you use more than one of their IDEs on a regular basis, I'm going to be honest - it starts to make more sense that the change between IDEs is less jarring. I use Rider a lot, and Pycharm and Android Studio infrequently, and I like that all of the IDEs are so similar that I know how to do most things as the settings and menu system is almost identical. I also use the IDEs on Windows, Mac and sometimes Linux - and the UI is similar enough that I do not ever struggle to make something work that I have previously done in any of the platforms or OS versions
JetBrains IDEs always had difficult and unintuitive default keyboard shortcuts. I think it's just because of its age (same is true with Visual Studio). Whereas VS Code's out of the box shortcuts were relatively great. If JetBrains IDEs shipped with VSCode-style shortcuts as an option (not necessarily default) that I could switch to without having to manually remap everything, I'd be so glad to use them.
This is amazing news. As a long time IntelliJ user (Android dev for 7+ years) who wants to start doing more c++, having free access to Clion is a godsend.
One thing I wish JetBrains would explore is building a fully native NON-JRE text editor that is a direct competitor to Sublime Text, but has a rich plugin API. I would love to see their take on a Sublime like editor. Atom and VS Code were the only serious Sublime Text competitors for a while, but now VS Code has gotten insanely bloated over the years, and also Atom is defunct unfortunately.
I would love to see them build a fully native editor with their decades of knowledge.
They have Fleet which was supposed to be a competitor to VSCode. AFAIK it's written in Kotlin so still JRE (unless they are able to compile it to native?).
This is good news. I for one, won't change my toolchain, but having a "free for personal use" alternative is always better, esp. when the license includes open source development.
Like it or not, C++ is not going anywhere in the short and long term, so it's always good to have real IDEs around (CLion, Eclipse CDT, etc.) which can integrate with good instrumentation and give real time feedback on your code.
That's interesting. This led me to their page, which also touted the now-free WebStorm.
Can anyone weigh in on WebStorm vs. VS Code for JS/TS development? I'm developing a back end with Deno running locally, and VS Code has been decent for debugging and using the language runtime. Would WebStorm offer any advantages?
Just curious, for low level mid-size projects ON LINUX, like a simple OS or a compiler, is Clion over-complicated or suitable? And how does the debugging look like?
I'm using VSCode with a manually written Makefile, and all of my debugging lives in gdb tui mode in a separate terminal. I do prefer a better UI though, but right now it's fine.
One concern is that Jetbriain IDEs usually takes a lot of memory. I do have a 16GB laptop though, so should be fine.
This is cool, I briefly looked at using this to work on an unreal project, does anyone have any experience with using CLion with unreal? Is Rider or VScode still ideal?
The follow-on [rhetorical] question is what value it provides to you outside of its integration with the IDE (e.g. click to open, it being integrated right into the window of your editor, etc)
CLion (and other JetBrains IDEs for that matter) doesn’t send any of your code to AI in the Cloud (unless you use sth like AI Assistant yourself, of course)
All duckduckgo software is filled with telemetry and analytics that cannot be disabled (The search engine, the Android web browser, the Windows web browser).
Agree that it is a crazy resource hog and I don't like the AI integrations either. But I gotta say for UI and features it is very good. I haven't been able to get VSCode or neovim to offer the same kind of easily actionable code diagnostics and suggestions as JetBrains IDEs do out of the box. I'm not saying it's not possible, but I couldn't quite match it even with a fair bit of tinkering.
I like the AI integrations (they me use Claude Sonnet 3.7, Gemini 2.5 and GPT 4.1 with my all products pack) and quite enjoy their Junie tool, much better than my attempts at getting Aider working.
UI seems pretty okay, at least on the 2025 versions of the tools (in compact mode, Inter 12 as the custom UI font on a 1080p monitor) but still quite the resource hog.
Oh well, I’m actually going to try their Fleet as well after reinstalling my OS because it was worse than VSC the last time I tried it, might be better now.
It's not. The functionality gap between VSC and Fleet has further widened, and its future looks even more unsure now they backed out the Kotlin native stuff back to the main IDEs.
The community edition of intellij and pycharm is Apache 2.0 licensed and you can use those for commercial development. Pycharm is about similar to vs code (I've used both). But intellij blows the Java & Kotlin support in vs code out of the water. It's not even remotely close. Reason, Java has several decent OSS IDEs as an alternative and those would win by default if the only option was a paid IDE. Clion and Rustover are somewhere in between I suspect.
VS Code plugins vary from language to language. And commercial non OSS tools for native development are pretty common (e.g. Visual Studio, XCode, etc.). So, I guess Jetbrains feels more comfortable charging for Clion and Rustover for commercial development because they know they are that good. But nothing wrong if you don't appreciate what is on offer.
Yes these tools use some memory and CPU. But then I use a decent laptop as well so it doesn't matter to me. Pretty normal to be spending on proper tools if you do this stuff professionally.
You and I have had vastly different experiences. PyCharm, like the rest of their tooling, will catch the most amazing bugs and LSPs don't hold a candle to that
I’m intrigued—what’s your setup? Do you use any advanced refactoring features like Change Signature or Extract Selected Members? It’s been a couple of years since I last tried setting up Java in VS Code, and I’m curious how far things have come. Intuitively, I’d expect JetBrains to have the upper hand in this area, since they can build specialized UIs for complex tasks, rather than being constrained by the limitations of the Language Server Protocol.
Incorrect, many of them do, and make the determination that a few nice to have features doesn't outweigh the cost, lock in, and incongruence with the use of open source software where possible.
I know several world class developers who have invented corner stone technologies that use text editors without any plugins and just run the compiler in a separate window. It turns out that you actually are not held back by not having your hand held by an IDE, it was just a skill issue.
Nowadays I just use an LLM. None of these features are relevant anymore. Unless JetBrains has a proprietary LLM that is better at coding than what ever else is on the market (they don't) then there's no reason to pay money for their products.
I am not even particularly bullish on AI, but not seeing how LLMs have made IDEs irrelevant is like using Vim and crying foul about IDEs without trying one out.
My example, my coworkers use VS code, so I review their code and my IDE is highlighting a few bugs in a function, bugs like $x is not defined , so I ask if thir IDE does not show any warning for that code, and yes, VS Code had ZERO warnings. The dev copy pasted some code but forgot this $x variable in, either he needed to copy it or remove it.
So I am sure someone will say that after you install VS code you should install X and Y plugins or some npm packages that you trigger later in the build to catch this errors. With Intellij I get them without screwing around with VS code plugins or vim plugins, and I also feel good that I pay some developers to work on a tool I use then use Microsoft product that would instantly fuck me over when Intellij would be killed by this unfair competition. Or did I read recently Microsoft already started with their bullshit related to the extensions and AI ?
I generally agree with the sentiment—if something is free, there’s often a tradeoff. But when there’s a paid tier, the free version can act more like an entry point or hook to get users into the ecosystem, rather than relying on harvesting user data. In JetBrains’ case, broad adoption brings a lot of strategic value on its own (like establishing industry standards or building community mindshare), so it makes sense for them to offer a genuinely free version without necessarily treating users as the product.
This AUP is related to the optional backend services run by Mozilla. The frontends for all of these services are open source, with no usage restrictions. It doesn't affect how Firefox is licensed in terms of copyright. Additionally, the backends for some of these services (such as Firefox Sync and Mozilla Accounts) are fully open source, so you could avoid the AUP if you wanted to.
reply