We used CloudEvents internally at Segment as an evolution of the existing ad hoc format that evolved naturally over time. We were generally happy with its ability to bring some semblance of order, documentation, and guidance to our event format while also being flexible to changes. For example, we layered on the ability for internal services to return 207 Multi-Status responses to batched events and it didn't require major hacks.
Not the OP, but there even used to be a rubydramas.com site. Not longer available and archive.org doesn't work for me there, but there have been a few HN submissions about this.
It's been 13-15 years, Ruby-world was full of drama between DHH-fanatics, and their opposition. I won't be scavenging through Google's enshittified search to substantiate anything just to please the desire of random HN users.
And just to be very clear: I can do whatever I want (under what I believe is a good internal moral guidance); I wanted to comment on a vague recollection I had of the state of the world at the time and I did.
Given you explicitly marked it as 'vague memory' so people could choose to weight (or if they so wish discount entirely) your comment based on that, I don't really understand the complaint.
WarpStream relies on a proprietary metadata store hosted within their internal network to operate, so it's pretty unlikely that Jepsen tests could cover that.
If you're ok with the externally hosted metadata stores as well as the high per-request latencies (p99 of 400ms, according to WarpStream), it's highly likely that things like liveness and safety properties are pretty far from your mind. So, I wouldn't bank on them submitting to a Jepsen test. :)
I think that WarpStream relying on a proprietary metadata store isn't an issue for Jepsen tests. If I understand correctly, Jepsen tests treat the distributed databases (or logs like Kafka) pretty much as a black box. Jepsen tests introduce partitions and look for missing/unexpected items against the items that were acknowledged as written successfully by the system.
If you look at Kyle's blog post, https://aphyr.com/posts/293-jepsen-kafka, there is no mention of looking into a broker's storage or any storage for that matter.
FWIW we subject WarpStream to continuous chaos/fault injection in our integration tests and staging environment to verify correctness and liveness properties. I wouldn't say they're far from our mind, we've just made a big trade off around latency that we think will make sense for a lot of people.
Not sure how widespread this advice is, but my son’s pediatrician had us start introducing peanut butter before he was eating solids by adding progressively more peanut butter to his bottle. So it feels like this is becoming the recommended approach to avoiding food allergies.
It’s become more widespread guidance in the last decade. There are a few studies iirc comparing cousins living in the UK as opposed to Israel. The Israeli kids get a peanut based cracker as a common snack and have much lower incidence of peanut allergies.
I think in the US the issue is that there isn’t a ton of published material on it. My info may be out of date as my kids are well past this stage.
One thing to be careful about here, and I don't have any data or skin in the game here, is survivorship bias. It would be worth noting the difference in deaths from anaphylactic shock in very early childhood by country.
I guess what I'm getting at is that you might not be counted as having a peanut allergy if you're dead.
Where I live (The Netherlands) we had multiple child care professionals recommend we feed the baby both egg and peanut butter as early as possible specifically to prevent allergies as well, so I don't think this is controversial in the medical community at least.
In the US 25% of mothers are back at work within 2 weeks of childbirth (https://www.vox.com/2015/8/21/9188343/maternity-leave-united...). FMLA requires jobs to let mothers take time but does not require they be paid, and most folks are living paycheck to paycheck and need the money.
It's a Department of Labor study, so presumably done on people already in the labor force. Doesn't change how terrible the stat is, even assuming 50% of mothers aren't in the labor pool that'd still be 12% of mothers back in work in under two weeks.
Now that I've recently become a parent, it's... it's shocking. The idea of handing someone that young and helpless off to strangers raises a strong "NO!" reaction, right from the gut instead of the brain.
In my neck of the woods we get 1.5 years maternity leave, and experiencing it myself I'm honestly not sure even that is enough.
Strong agree. My son is nine months old now and if we had to put him in day care tomorrow, I wouldn’t mind. It takes a village and all that. Happy to delay that a little while longer, though.
Decidedly not. There's plenty of meals given to kids outside of school to perform this intervention, it doesn't need to be in every meal on top of what other people have said about the timing.
It's unfortunate that their reputation took such a massive hit with the public profile debacle. The folks I talked to / worked with at Triplebyte were all great people with their hearts in the right place. It's just an incredibly tough market to be in.
I think the problem was that what they were trying to do couldn't scale. As I said I interviewed in the early stages with Ammon and it was a great experience. He knew what he was asking and he knew how to evaluate your skills. But how can you have a hundred Ammons interviewing and assessing people? It just doesn't scale.
Interviewers actually did scale reasonably well, which is part of why Karat (which stayed focused on that part) was in a position to buy our stuff to begin with.
What didn't scale was the ability to get high-quality candidates in the door in the first place.
6 years ago, I did Triplebyte as a candidate, got 6 interviews at good companies and 4 fantastic offers (including one from them, directly!). I ended up accepting an offer for a company that had a life-changing exit for me, and it wouldn't have happened without Triplebyte. So I have a lot of love for them. It was sad to see them stumble with the weird public profile stuff, but I wish only the best for them. I hope this was a good outcome for the folks that I talked to all those years ago.
Nearly same journey for me. I can’t thank them enough for organizing everything (including the flights, hotel, and scheduling) and putting me in front of several great companies as someone who at the time did not live anywhere near the bay or have easy access to the SF startup scene. It felt like a really premium experience, and one that I’ll forever be grateful for.
These kinds of stories were really motivating for all of us who worked here at the time, and they definitely soften the blow of the final result, at least for me (and I think for Ammon as well).
To give a different perspective, I applied to Triplebyte when they announced their project track for people who don't do well in interviews. They told me I had done a great job on the project, but that they couldn't move me forward because they needed candidates who could do well in interviews.
(What was the project track for??)
They recommended practicing on interviewing.io, a site with closed membership that kept me waitlisted for over a year (and only then let me in because I complained about the situation on HN, not because I'd actually waited long enough to get off the waitlist).
So as far as I could see, Triplebyte advertised one thing, delivered the opposite at every opportunity (I'm also still bitter that the site with the tagline "No resumes, just show us you can code" opened its interviews with "So, where have you worked in the past, and what did you do there?"), and didn't even pretend to be interested enough to give you advice that was, theoretically, possible to follow.
Same here. Triplebyte quite literally changed the trajectory of my life. I wasn't really aware of how things have changed over the past 6+ years as mentioned in the other comments, but I'm sad to read them. I wish the whole team well, regardless. <3
News sites have long pushed for regulations that would require companies like Google and Facebook to pay for excerpting and linking news articles, so part of me wonders if this is an intentional policy to avoid being held liable for paying, in turn, their sources of information should those regulations ever come to pass. Deeply funny to me, honestly.
I think Guywithabike is saying that phys.org doesn't want to have to pay Greenpeace for citing them, should that future come to pass of Google paying news sources for snippets.
So I'm a big dummy, but I STILL don't understand how current stock price affects companies. They've already raised their money. How does the stock going down make them lose money other than lowering their chance at another seed round?
the stock price is a trailing indicator of the company’s health. in the case of headcount, you hire based on growth rate, to sustain and even further that growth. it works until it doesn’t. then you fire.
it’s no skin off the company’s back, so it is the correct strategy. when CEOs (the world over) say they overhired AND that it was a mistake, it’s a pitiful political lie. also a mandatory strategy. you can’t say, “employees are cogs. we planned this shit. this was plan A.”. you wouldn’t be able to hire again when you find your stride.
Hotline was full of warez and such, but it was also home to a number of vibrant, welcoming communities. I was a little bit late to the BBS heyday, but Hotline served that purpose for me and for many others. There was one called REALbasic Cafė centered around a commercial cross-platform programming language that I spent a lot of time it. I even attended conferences with other members, flying across the country alone as a high school kid. I have nothing but fond memories of Hotline.
