How did Replit respond to this blog post?
intuitiveexplanations.com
intuitiveexplanations.com
I was reading through the replit blog to see what kind of things are trade secret vs public, and noticed that they didn't go out of their way to highlight that Bret Victor's "Learnable Programming" is clearly the complete inspiration for replit's debugger (time travel, no hidden state, etc). They have a small link to Bret's work but don't mention him by name. Considering Bret's article is basically THE piece to read w.r.t. programming environments, and the gif that replit shows looks pretty identical, it seems right to have his name/article clearly at the top.
"you said 'I will not be copying more stuff' " (OP didn't say that and given their current stance of not copying anything, I doubt they would admit that)
"Arrive at new ways of doing things -- make new mistakes"
Egotistical people always try to get the last word in. Even when apologizing.
Like a used car salesman sliding in some hidden fees after the deal is done. All of this is poorly saving face anyhow.
"Hey, really cool project. Though honestly, not sure how I feel about a former intern of mine publishing a project that is so similar to what they worked on as an intern. But on that note, we'd be happy to have you back on the team if you want to continue the project with us :) What do you say?"
or something along those lines. I still wouldn't agree with him, but at least I'd sympathize.
instead he went full-on attack mode, lawyers and all. sheesh. talk about tone deaf.
everyone is armchair quarterbacking here on what was the rational thing to do. but we are mostly not rational creatures esp when our "babies" are involved.
Even now Amjad refuses to provide any detail or receipts to Radon for his public accusations of tortious behavior.
> everyone is armchair quarterbacking here on what was the rational thing to do.
Do you seriously prefer your web psychoanalysis?
The difference here is that CEOs are expected to handle things professionally, especially CEOs of a VC-backed startup.
As noted in the previous thread, the "rough day" excuse doesn't fly when you double down and offer legal threats.
I only ask because there is a 3-page (in my opinion, aggressively worded) email response in between "I feel this is unethical" and "legal action". I could see how the content and tone of that particular email could be interpreted in a way that could escalate the situation.
Again, I do not agree with the response here, but it's not as if there wasn't a long email with a fair amount of context in the middle.
But if you are going to make statements like "I can't believe he responded in that way!!", it's important to read the thing that is actually being responded to.
The response was written by a CEO. C level positions carry a certain duty and require tremendous skill. In this case, Amjad responded to a poorly written email in such a way that every investor and employee has to fear or they’re not paying attention.
Humans have thousands of years of experience processing trust. It takes a life to build a good reputation and one email chain to destroy it. C level leaders genuinely have to be a lot better than this, especially when faced with difficult emails.
Taken directly from the email; "but this feels weird/unethical since you worked for us and now are copying internal design decisions (even bad ones)".
Could Radon's email be worded better to defuse the situation? Probably; but they are a fresh graduate just out of university and I think it's reasonable to hold Amjad to a higher standard.
So much egotism from the CEO in this exchange. I'm glad this was published and posted to HN because it's a valuable lesson to learn by example from.
We have the public commit history of the project. Feel free to go through each commit and point out what was stolen from you.
I'm not saying that I agree with threatening legal action here, but this isn't really a fair classification. You are implying the the CEO put an ex-intern on blast in the community. There is a huge difference between making a statement in a private email (which the other party then releases), and making a public accusation (which is what you're implying).
These are the words Masad used on Twitter to describe the situation in the OP:
> As a matter of principle, when someone goes into your home and steals from you, even if it's not material, you have to respond.
That's a direct public accusation of IP theft.
Problem is, using a web-based terminal to access a host is not original. They have no patent on it.
In general, such an accusation could be legally actionable. E.g. say you were employed by some restaurant and stole their secret recipe, and then used it to offer the same menu item in your own restaurant. Type of thing.
There is no secret sauce here, though.
The whole approach he took here is just very poorly thought out.
Though yes, he seems to have some personality flaws. I hope he's working on them, but also don't think he should get a pass just for admitting them.
I think the point the founder was probably getting at was to recognize that he is no longer in a underdog / position of no power / have no resources (aka struggling kid from Jordan) but is now in a position of power and needs to reset his mindset and no longer have the 'do anything to win' mentality that it often takes to get out of poverty / fight as David against Goliath, etc.
Not sure why you would think he is implying anything about people from Jordan not being kind rather than just state you don't get what he's trying to get at.
The main point is that he apologized and undid his mistake, what else do you want?
Also he's obviously not saying that people from Jordan can't be kind, he's saying that his upbringing may have had something to do with it which is completely possible.
I really don't know what else do you expect, I find your comment unnecessarily negative for an apology that came with a redeeming action.
This is something that’s true, but not really relevant.
Note that we did not reference all of our past traumas these two comments. Not because our traumas didn’t shape us, but because it’s not practical to describe our traumas in every thing we say, all the time.
I just kind of ignore it and assume it’s a natural rhetorical technique, sometimes unconscious, to just say something that’s perceived positively. I suppose technically it’s “appeal to irrelevant authority.”
But, and this is important, just stop. For five minutes. When something upsets you, and you JUST WANT TO SEND THIS EMAIL - don’t. Wait five minutes. Take a walk. Tell someone else about the issue.
Then, and only then, after you’ve not thought about the email for a bit, erase it and write it from scratch.
It helps. It’s therapeutic. You don’t end up with your emails screenshotted and posted to the top of HN for two days in a row.
Assume the best intentions. I find far fewer people are looking to take advantage of you than are simply looking to be liked.
This would solve a lot of problems, I think.
Emm. Really? Since when does he hold such a power?
Everyone hates it, but this is why you involve lawyers from the start if possible. It's better for everyone in the long run.
If you find the quote, let me know! I’m always looking to expand my mountain of unread books...:)
The top YouTube comment(s) say the quote I was thinking of: “You can’t give me my rights, I was born with them. You can take them away, but you can’t give them to me.”
I think that the situation has escalated though partly due to the approach of the former intern here, who - while rightly calling out the non-apology aspects of the apology is aggressive themselves now that the power is with them courtesy of the public exposure and general backing of HN.
The blogpost here drilling down into every perceived slight and human failing lacks grace and magnanimity and doesn't seem much better at handling the favourable side of the power balance than Replit did in the original correspondence.
Maybe time to move on.
The CEO Amjad Masad was still doubling down on claims that Radon is a thief on Twitter even after he apologized. And he still refuses to provide details or receipts to his public accusations, despite Radon's willingness to provide receipts.
But you are asking for Radon to be gracious?
I think it's fair to forgive some lack of grace. Sure, the post could be more generous, but the whole handling could have been, and the poster wasn't the one making the legal threats.
You forget that the author is personally offended too. He just has no way to extract a proper apology because he's an unemployed fresh graduate up against a well-connected CEO. It's not even about legality anymore, since there's technically no wrongdoing here. It's just a sincere apology from one human being to another, but Amjad's never going to give it because it's pretty clear from this exchange what kind of person he is.
EDIT:
Just to clarify. This post has gained over 200 points and 70 comments on the last hour. It jumped from top 5 posts to low end of the top 50 posts in seconds and is rapidly sinking. Just curious what decision led to the hiding of a popular post.
However, we moderate HN less, not more, when YC-funded startups are involved in a story (https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...). In keeping with that principle, we took a hands-off approach with yesterday's thread.
Today's post, however, is a follow-up to yesterday's. Indeed it's basically the same story. There are well-established moderation rules about that: repetition is bad for curiosity [1]; we downweight follow-ups [2] unless they add significant new information (SNI), defined as information capable of supporting a substantively different discussion [3].
Moderating HN less does not mean not moderating it at all—that would be a gigantic loophole. Having bent the rules drastically to accommodate yesterday's thread, I don't think it would serve the community to do the same thing all over again the next day. Indeed, we've gotten a significant number of emails about this post, asking us why it was on the front page when the same story was so prominent yesterday. That's because HN users are used to the site not being overly repetitive. People come to HN for new and above all different things (diffs are what's interesting [4]), and we need to take care to preserve that property.
[1] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...
[2] https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
[3] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...
[4] https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...
Edit: I posted a related comment at https://news.ycombinator.com/item?id=27439412.
album: https://imgur.com/a/QcHG9VK
In the first screenshot, notice that the highlighted comment has no replies. Here's a link to the page I was on, anchored to that comment: https://news.ycombinator.com/item?p=3&id=27424195#27429786
In the second screenshot, after navigating to the comment directly, there are in fact replies: https://news.ycombinator.com/item?id=27429786
The bug is that those replies should be showing up in the first link, shouldn't they? E.g. https://news.ycombinator.com/item?id=27430242 is a reply to https://news.ycombinator.com/item?id=27429786, but when viewed on the original page https://news.ycombinator.com/item?p=3&id=27424195#27429786, it's absent.
I know that "more" pages shift toplevel comments around, but I've never seen child subtrees get completely pruned with no indication that they exist at all unless you navigate to the comment directly.
Sorry for the distraction. I didn't want to email this for a few reasons, some of which might be invalid. However, it does seem to be a slightly-serious bug if true, since a recent refactoring must've caused deeply-nested children not to show up when browsing top-level "more" links. I'm not sure anyone else has noticed yet.
(If this is intended behavior, sorry!)
Good luck bug hunting.
We're experimenting, in order to deal with this problem: https://news.ycombinator.com/item?id=27437362.
(Wish it was a different color, or included the descendant count.)
Ideas about how to get this right are welcome.
seems intuitive to piggyback on the same fading ui hint as comments, having a bright "more" indicating standing of hidden comments...
14 years was a really good run. I suppose even lisp has its limits. :(
Yeah, adding the descendant count would make it pretty good. Hard to think of anything better.
I’ll be honest: I’m really worried this will become another tool for moderators to control discussions, while concealing the fact that it was a moderator (rather than algorithmic) decision. But no one can stop you if that’s the kind of control you want for your team.
The flip side of that is, since 2018, your team has done a stellar job, as far as I can tell. I don’t know if something changed, but things seem calmer now. I can’t even recall the last worrisome moderator decision, and I’ve found myself arguing on your behalf more than once.
Dunno. You’ll get it right; you usually do. Thanks for being welcoming about ideas, and good luck wrestling with transpilers and performance issues.
Or just open your browser console and run the JS directly:
[...byTag(document.body, 'a')].filter(x => x.innerText == 'more').map(x => { x.innerText = `${byClass($(x.href.split('item?id=')[1]), 'togg')[0].attributes['n'].value - 1} more`; });
I find it much easier to browse HN now. It's immediately clear whether I'm missing a large subtree.From analyzing it today, one alarming thing I noticed was that in all recent threads with "more" links, the conversation never seemed to nest more than one or two levels deep. But the moment you go to an old thread, e.g. https://news.ycombinator.com/item?id=21231208#21232605, the "more" links turned out to be hiding 9+ comments.
But, like, it's not a horrible thing to constrain the comment subtree. If you were feeling like it was the way to go, my only feedback is to trust your judgement on that one. Personally, I thought it was an interesting new dynamic, even if I had my own (probably-unjustified) concerns.
Anyway, best of luck, sorry for the mistaken bug report, and keep trying new experiments! It's really cool!
Edit: not responding to guilt but to shame.
Personalities like this exist across the social stratum. They are something to take careful note of as quickly as possible and avoid.
And unwilling to backup their public accusations with detail or receipts.
The CEO was also on Twitter doubling down on the accusation that Radon is a thief even after posting on HN.
Replit used legal threats to kill my open-source project - https://news.ycombinator.com/item?id=27424195 - June 2021 (1234 comments)
(Edit: I'm adding some text here because people have been linking to this comment.)
This story was at #1 all day yesterday, in contradiction to HN's normal rules. That is, we bent over backwards to allow this story to have much more attention on HN than it would have gotten in a normal case. That's why it got 4000 upvotes and 1300+ comments and was the #1 story the entire day (https://news.ycombinator.com/front?day=2021-06-07).
Why did we do that? Because we moderate HN less, not more, when YC or a YC startup is the story: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu.... That is literally the first principle of HN moderation, and we stick to it strictly.
Less moderation does not mean no moderation, however. It does not serve the community, or the quality of HN, to repeat the same thing all over again the next day. Not having the same story repeat itself two days in a row is standard HN moderation.
Downweighting follow-ups in order to avoid repetition, is a bedrock principle of HN; otherwise the front page would be dominated by the same few stories every day. There are lots of past explanations about that at these links:
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...
We bent the rules considerably to allow the story to keep a prominent position it would not otherwise have had, for an entire day. I don't think it would be proper to do the same thing two days in a row, and indeed many users were emailing us to complain, when they thought that was about to happen.
Masad, if you're reading this, get off your high horse. Bullying and scaring people into compliance with lawyers does work, but you pay it dearly in PR, especially when you're working on products oriented towards highly opinionated people.
Really, this was a great opportunity to acknowledge and learn from aggressive mistakes while still running just a series A company. It was a blessing to the CEO to be highlighted now before the real weighty damage occurs. Unfortunately, it's pretty apparent to most of us and future investors that his public persona will get eviscerated if his company proceeds to ever become more in the public spotlight.
Not when someone can banish it from the front page[1]. See this comment timestamp[2] and compare it with the sudden drop in ranking on the upvote tracker chart.
We bent the rules considerably to allow this, for reasons I've explained at length here:
https://news.ycombinator.com/item?id=27439301
https://news.ycombinator.com/item?id=27439412
https://news.ycombinator.com/item?id=27427198
https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
Not having the same story repeat itself two days in a row is standard HN moderation.
p.s. Please see https://news.ycombinator.com/item?id=27433981. If you could please stop creating accounts to break HN's rules with, we'd really appreciate it.
It's very important to understand that this story was at #1 all day yesterday, in contradiction to HN's normal rules. That is, we bent over backwards to allow this story to have much more attention on HN than it would have gotten in a normal case. That's why it got 4000 upvotes and 1300+ comments and was the #1 story the entire day (https://news.ycombinator.com/front?day=2021-06-07).
Why did we do that? Because we moderate HN less, not more, when YC or a YC startup is the story: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu.... That is literally the first principle of HN moderation, and we stick to it strictly.
Less moderation does not mean no moderation, however. It does not serve the community, or the quality of HN, to repeat the same thing all over again the next day. Downweighting follow-ups in order to avoid repetition, is another bedrock principle of HN moderation; otherwise the front page would be dominated by the same few stories every day. There are lots of past explanations about that at these links:
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&sor...
https://hn.algolia.com/?dateRange=all&page=0&prefix=false&so...
Amjad, if your securing your moat via threats, you better look out.
To anyone else, go grab Theia. In about an hour you can get something better than Replit going, and if you combine with Nix/Guix, there's just no reason to deal with people that behave like Amjad.
Kinda the same thing here
I think I see one part of the problem, though. Many people implicitly believe the claim about Riju copying repl.it is itself ridiculous. So I want to talk about this.
It's famously said that execution is worth more than ideas. Meaning a large part of a company's value is in its acquired knowledge from iterating on designs and the paths they landed on. Imagine everyone who sets out to create an online code runner starts at the same point in the online-code-runner problem space. Every decision you make is uncharted territory, and every choice, small UI changes, how you serialize messages, how you orchestrate containers, has consequences.
When you take the result of those learnings -- a very specific design path -- and copy it, it can seem like you were just doing the obvious thing, precisely because the first company carved that path first. It seems like nothing, and you'll never enforce it in law. But what you're doing is basically pirating the encoded knowledge from all those iterations and trial-and-error cycles; instead of starting from the initial point in the problem space and independently reasoning at each point what direction to "step" in, you teleport right to the same spot as the company, getting for free any progress they've made.
Saying there are a bunch of repl.it clones, etc. misses the point. The "online code runner" idea is cheap. That's not what Amjad is saying. It's not even about individually patentable trade secrets. It's more like, we've been iterating in this problem space for a long time, an intern comes in, sees the path we took, leaves and releases something that takes largely the same path, then tries to claim the path is just "the obvious route." But anyone who's worked on a long, hard problem knows that's not how it works. A million micro design decisions go into eventually arriving at the path you take. It's all the more suspicious if they made all the same micro-mistakes you did (as Amjad claims). What are the odds they started at the initial point, independently reasoned through each sub-decision, and made the same "steps" in the problem space you did, mistakes intact?
It may not be legally enforceable, but I think it's very morally shady.
I'm not saying Riju did copy repl.it, by the way, just that it's not a completely crazy accusation. And I don't think Radon's blog post definitively proves he didn't. It goes through his tech stack in a very accounting-like, line-by-line breakdown. But it doesn't address my point above, the point I think (can't speak for him) Amjad was trying to make.
So if Amjad saying "I still think you crossed a line ethically, but I apologize for reacting by making legal threats and abusing power" -- which is, actually, the entire thing he did wrong -- strikes you as insincere, what's the right way for him to assert his own moral grievance here before this whole matter wraps up?
The reaction to the situation from the CEO and the reaction to the reaction by the development community are both soaked in cultural faux pas generated out of a few decades of startup-pop-goes-the-weasel.
I’m sure both sides love the untrue misnomer phrase:
It’s not personal, it’s business.
No one owes Replit success. It doesn’t matter what the CEO says, I’ll be using alternatives. There’s really nothing to “wrap up”. Other than to act as a cautionary tale to other executives who may be tempted to flex a power differential.
"I'm sorry I slugged you but you made me really mad. I won't do it again if you don't" is wholesale failure.
Sure, Amjad should apologize, and he did. Shouldn't Radon apologize? Why is this a completely one-sided thing? Why is Radon allowed to be like, I don't think I did anything wrong, and I'm disappointed by Amjad's behavior, but Amjad isn't? In this hypothetical, they both did something wrong, shouldn't there be a mutual apology?
Not complicated, really.
Good that Radon has responded well to that criticism and made communication transparent in his blog.
Setting the wrong tone from the position of CEO.
“Virtue is what you do when nobody is looking. The rest is marketing”. – Nassim Taleb
I'm sure from the information at their disposal, the CEO thinks their IP was stolen, and feels obligated to defend it vigorously. Yes, he did not provide specific claims, but often it considered unwise to be specific in these situations (both when threatening legal action and when backing down). What's described as "misquoting" could be deliberate, but given how all of this seems to be in the context of two parties not communicating effectively, it is far more likely a case where what was heard from the CEO's context was very different from what Radon intended to communicate.
Like I said, I don't begrudge Radon's feelings and perceptions of the situation, and he may well be perceiving things accurately, but with disagreements like this, the norm is that all parties have a distorted perception. I'm just glad it has been resolved without going to court.
Never, ever, EVER release an open-source version of your current or former employer's product. If you do, the employer can establish you had access to their code and they can argue that your work is substantially similar to theirs, meaning THEY CAN AND WILL SUE YOU for copyright infringement.