JetBrains defends removal of negative reviews for unpopular AI Assistant
devclass.com
devclass.com
That shouldn't be considered a valid reason to remove a review. I could maybe understand down-weighting reviews as they age and as issues are resolved, but as a potential buyer of some product/service/whatever, knowing that something was released with a bunch of issues (even if now solved) is a valuable signal. Preferably, they would reply to reviews and say "XYZ was addressed in update ABC" or something.
Nuking reviews is a valuable signal as well, I guess. Just not in the way that they hope. Knowing that they've done that has (further) lowered my impression of them.
But this needs to be transparent - hidden average score manipulation just leads to distrust in those scores.
"We fixed yesterday's outage, review removed"
That is more fair than Jetbrains removing other people's negative reviews unilaterally, especially when it remains to be seen whether Jetbrains actually fixed the issues.
Legally, Jetbrains is allowed to be judge, jury, and executioner of reviews on their own site. Morally, they should not act unilaterally, and should gain customer approval before harming customers.
> If you handle critic and improve
You are assuming they improved. With the reviews deleted, there is no way to know for sure, they could just as easily have gotten worse and hidden it by deleting the review. That's why you don't delete the review.
The proper move here would indeed be to post a reply, and have a public conversation. Readers of reviews can then read the reply, and either trust them at their word that they fixed it with no bugs or other mistakes, or test for themselves.
tl;dr: the reviewer, not the biased company being reviewed, is the judge of whether the issues described in the review have been addressed / fixed / made irrelevant. YES, there may be downsides to this, but they are not as bad as the downsides of the alternative (company deletes arbitrary reviews about themselves for arbitrary reasons with no oversight).
“ I previously submitted a review critiquing this plugin, but it was removed by JetBrains moderation — an unfortunate decision that, in my view, undermines trust in open feedback. I have now tested the latest AI plugin (v243.23654.270.16). The plugin does offer limited support for third-party providers like Ollama and LM Studio (the latter being a better fit for most local LLM users). However, this support is restricted to chat interactions only — not to autocomplete, inline suggestions, or in-editor refactoring tools. In practice, this limitation significantly reduces the plugin’s value for users who already maintain ChatGPT Pro accounts or local LLM workflows. Rather than fully enabling local model integration, the design seems oriented toward promoting JetBrains’ proprietary cloud models and subscription services. Specific ratings: • Integration with IDE: 5 stars — Excellent UI integration into JetBrains products, smooth setup. • Performance: 1 star — Noticeable latency compared to local models; frequent delays. • Available Features: 1 star — Limited flexibility for serious LLM users; core features locked to cloud services. • User Interface: 1 star — Chat feels bolted-on rather than deeply native; inconsistent UX across project types. • Documentation Quality: 1 star — The documentation exists but feels sparse, with limited guidance on third-party setup and unclear disclosures about feature limitations. While some users may find the plugin sufficient for lightweight AI chat, in my assessment, it falls short both in technical flexibility and in respecting user choice. Thank you to JetBrains for providing the opportunity to share my neutral and unbiased observations with fellow developers” [1].
[1] https://plugins.jetbrains.com/plugin/22282-jetbrains-ai-assi...
They seem to have to say that a lot about this product, yet they don't really seem to learn any lessons. When the original flood of bad reviews came it it was because they made that plugin bundled with the IDE and then had a "bug" where it couldn't be effectively removed. There was no precedent for bundled a paid plugin nor need for it to be bundled with the IDE. Just their desperation to cash in. They then walked that back with the same "we could have done better".
This is more of the same. The "AI Assistant" still lives on the default side bar regardless if you have that plugin installed or not.
At this point, they know they could do better yet are choosing not to.
> There was no precedent for bundled a paid plugin nor need for it to be bundled with the IDE. Just their desperation to cash in.
Not only are they quite open it is a means to cash in more InteliJ licenses, the community has generally an anti-Java sentiment, especially on Android, as if it doesn't depend on the ecosystem for its own existence.
The language might have a great design, however for me that is off-putting and makes me rather play with other JVM languages, like Clojure.
"The next thing is also fairly straightforward: we expect Kotlin to drive the sales of IntelliJ IDEA. We’re working on a new language, but we do not plan to replace the entire ecosystem of libraries that have been built for the JVM. So you’re likely to keep using Spring and Hibernate, or other similar frameworks, in your projects built with Kotlin. And while the development tools for Kotlin itself are going to be free and open-source, the support for the enterprise development frameworks and tools will remain part of IntelliJ IDEA Ultimate, the commercial version of the IDE. And of course the framework support will be fully integrated with Kotlin."
https://blog.jetbrains.com/kotlin/2011/08/why-jetbrains-need...
> The next thing is also fairly straightforward: we expect Kotlin to drive the sales of IntelliJ IDEA.
So I just went back to CoPilot.
I know they don't have deep pockets, but, like you, I'd rather they just spend it on making a good tool.
I used to be a huge evangelist for JetBrains products, I loved having a product where I felt like I could just pay and get something of quality, it's really sad seeing that devolve into the same mess of "you are the product" as virtually everything else, despite the fact they were still demanding my money.
Seriously. This is where Jetbrains went astray. Their original programmers I think were pretty high quality, but their product people are clearly so pants-on-head clueless that they wouldn't know a golden goose if it crushed them under metric tons of golden eggs. Good things don't power-law scale and "let's make 1000 forked IDEs and 50 really weird infrastructure servers" isn't a winning strategy. It's how you bury yourself in tech debt.
> I'm not sure Google really wants to make a public IDE.
I mean, that's what launched Jetbrains in the first place. Google needed proper development tools for Android and a language more modern than Java at the time to support it. So we got Android Studio and Kotlin. They dumped a bunch of money into them and this is what we got once they started to get more independent.
Vibe as a service
I spent about a month getting to grips with it, and now i feel more productive than ever, and I regret not doing it sooner!
Also I can rest assured that it will be forever free, no more annoying licenses to deal with.
Not sure why it’s so hard for them to catch up with Cursor. They have everything they need but somehow they focus on just something that they don’t have much expertise, building models instead of better integration. It’s a shame seeing such good product going downhill considering AI is becoming fundamental for dev productivity.
I am a lifetime user of PyCharm but the reality is that Cursor is just so much more productive now. “Junie” is a decent attempt but nowhere close to Cursor yet.
At least with code completion it's pretty obvious at this point that no one needs the overpowered top-line models, and the trajectory on local LLMs is such that I don't think it's unreasonable for them to hope to avoid the big players entirely.
They don't need to beat Claude for it to work, they just need to keep their customers satisfied.
Disagree, I keep trying Jetbrains once in a while and keep walking away disappointed (used to be a hardcore user). I use VS Code bc it is seamlessly polyglot. Jetbrains wants me to launch a whole separate IDE for different use cases, which is just horrible UX for me. Why would I pay hundreds for a worse UX?
you also have to install plugins in vscode
> And it's free
not relevant to my comment
Visual Studio is still the best for C++, followed by C++ Builder.
VSCode is the only Electon app that I tolerate on my own computers.
When is InteliJ finally going to support JNI development instead of requiring two IDE licences?
One of my two major gripes about Kotlin -- that, and the lack of a standardized style guide and formatting tool so that there's no bikeshedding or difference between repos.
Of course that is no Kotlin outside InteliJ, JetBrains even has a blog post on how Kotlin is supposed to help selling more licenses.
https://blog.jetbrains.com/kotlin/2011/08/why-jetbrains-need...
By the way, for a little while there were some JetBrains employees doing an Eclipse plugin, but naturally they ramped out that effort.
https://blog.jetbrains.com/kotlin/2015/09/kotlin-eclipse-plu...
I also believe they should really stay calm and not get sucked into the AI hype. Worst case they will be the heroes to the people who like to program for the joy of it, in case these AI IDEs should really take over (which i highly doubt).
Lately they've also been coddling with the VSCode crowd by aggressively pushing the new UI over loud objections of old loyal users.
Either one seems like direct opposite of the hacker user you're mentioning.
Having done go and python in jetbrains and vscode, I definitely enjoy the experience in jetbrains more. A lot of java people like IntelliJ for their Java and Kotlin support.
OTOH, copilot has been not as good on Jetbrains as it has been on vscode. Updates are delayed to give VSCode a first mover advantage to VSCode.
Google Gemini Code assist plugin last week still sucked, didn't try it today.
Copilot can also use Gemini Pro 2.5, but they delayed the release of the plugin for Jetbrains, and only have a context of 10 files I believe for the edit mode.
And I thought I read somewhere that Jetbrains AI Assistant can use gemini AI pro, it's limited to a context window of 200,000. I might be wrong on that.
Junie is reasonably good, but still has issues with understanding large code blocks of more than a couple of kilobytes. But it applies the changes first, without letting you do a review of the code. The only real way to do it, is to check in the code in git, then let it run, and then look at the results.
I've asked Junie to fix unit tests using brave mode, and it seems more than capable with that.
I think the trick with Junie is small defined tasks, rather than large bullet points. Or at least have a detailed plan which you can paste in, and reasonably detailed so it won't have to guess or infer what it is you want.
But generally speaking, I've had far better luck with Google Gemini Pro 2.5 on code generation than with some of the others lately.
Edited to add: Github Copilot added agent mode. I'm going to try it now.
- I go to its website and neither on the homepage nor the features tab does it bother listing what languages the IDE is even for. Is it Python? C? HTML? It's an IDE .. for what? What languages? What project types? How can they not list this basic fact?
- Oh well, click the big Download link, and it downloads an app image file. No idea what to do with this, never seen one before, have to google it.
- Mark the file as executable and run it and get a cryptic error: "The setuid sandbox is not running as root" and it errors out.
- Back to google, google for that error message. Find various Cusor bug reports and people complaining about it but they haven't bothered fixing it.
- Find a workaround, to pass in a –no-sandbox arg when running Cursor, and now I get it to launch.
- It opens up but the text is incredibly small on my (4K) monitor and the text coloring is a dark grey that's almost indistinguishable from the background color, immediately go look for settings to fix it. There's ~50 settings results for "font" or "size", I change a few of them and it seems to make no impact to the UI font and I quickly give up and just want to try the editor.
- I read online that I need a "CMake Tools" extension to open a CMake project. In cursor I open the extensions marketplace and search for "CMake" and there's zero results. I try to open a CMakeLists file anyways and it opens it as a text file and then prompts me to install a "CMake Tools" extension. Ok? Why didn't it show up in the marketplace before?
- I click the popup about the CMake Tools extension it opens the marketplace page for it, showing me the details about it. Whilst I'm reading the details for example to see who the author of the extension is, whether it's even a legit extension or not, the reviews of it, it just automatically installs it by default without me clicking the Install button that was on the page.
- After installing the extension the CMake file I opened is just in a tab but hasn't imported the CMake project, so I close it and re-open it from the File->Open menu.
- It again just opens the file as a plain text file and doesn't actually try to import the CMake project in any way, I don't see any popup or button or call to action to actually import the CMake project in any way.
- I give up and just switch back to my normal IDE
Half of your bullet points above are you being intentionally obtuse to be grumpy in this comment.
Among the ones you mentioned, I also tried Gemini Code Assist JetBrains extension, but it doesn't integrate anywhere close to what Cursor does. (Direct code inserts, rollbacks, checkpoints, context integration) Zzzzz come on JetBrains
“Hi, we’ve updated and these issues should be addressed now. Please take a look and let us know what you think!”
If you control both the product and the platform, deleting negative reviews is much more convenient than actually resolving the issues.
Have switched to my very old workflow of using nvim and customizing it with NvChad.
> AI services are expensive to provide, because they tend to be processor-intensive, but competition between vendors is a likely reason for JetBrains introducing a free tier earlier this month
If it's so expensive, why do they force it on everyone? Sure, a lot of folks want support for this, but enabling it by default is just annoying for their long-time users. Not to mention the costs of full AI-completion, I hope they don't get the idea of also enabling that by default.
Uh, I don't think it's ever been "lightning fast"... Great at refactoring, navigation, and boilerplate generation, yeah; fast, no.
Unless something drastically changes, I won't be renewing my license anymore. I don't like VS Code, but it's been much more reliable than the Jetbrains tools I use.
(Please don’t disappoint me by saying “they added an optional feature and I don’t like it”)
They have had quite a few issues open for nearly a decade (I wish I was kidding) for features that had been quite sought after, that free tools have had for years.
- DataGrip sporadically stops working when returning from sleep. I have to force-kill it to continue.
- Take a relatively empty file with 10 lines. WebStorm is supposed to reformat on save. It hangs for 10+ seconds or until I cancel reformatting.
- I saw a low memory alert for the first time in months this week. My workflow hasn't changed drastically and I wasn't running anything, just editing a few files.
- Overall everything feels a little slower than it did a couple weeks ago with the older version.
I don't think it's worth filing issues in YouTrack because I've seen those go nowhere in the past.
Another annoying thing is they also removed the modal commit window in favor of a VS code style commit. It was removed without notification, and I had to install a plug-in to restore it.
IDE UI is the most important thing for me, I've built muscle memory to use it without think. When they tinker with it, it forces me to think about the IDE instead of about what I'm working on, and that's really annoying. Not enough to lead me to change, for now.
I'd pay decent money for an editor that could be faster than Neovim but until then they get my yearly donation.
There's so much that can be gained solely on speed alone. Especially for JetBrains products here too.
Haven't it used it since adopting neovim but I am happy that the Sublime team is still around and able to make a living on a good product.
Junit test runs say No tests available about 50% of the time
It feels slower and slower
Today I started getting "unable to save settings" or something, no idea what that is about
It really shows that they're distracted from quality. My guess is with the breadth of features and quite amazing attention to detail, they needed 100% dedication to those efforts, and now a chunk of the company is doing something else, and now the product is falling apart
But hey we have a totally new console engine or something so that's really nice (I've personally never used the console ...)
In Webstorm it sometimes does not recognize comments (colorizes them as code), does not reliably recognize multi line ToDos, and frequently warns me in a if (myVar == null) that myVar may not have been initialized.
At least the last issue is as old as the hills, has been reported several times, and yet they seem to be unable to properly fix it.
For multi line ToDos, make sure the comment in the second line begins with a space.
> When you open a workspace with Augment enabled, your codebase will be automatically uploaded to Augment’s secure cloud.
[0] https://docs.augmentcode.com/jetbrains/setup-augment/workspa...
[0] https://www.ftc.gov/news-events/news/press-releases/2024/08/...
I would LOVE to see a paid plugin challenge Jet Brains on this by asking that their old negative reviews get deleted under threat of legal action.