rocmcd

100%. Context propagation is _the_ key to distributed tracing, otherwise you're only seeing one side of every transaction.

I was hoping odigos was language/runtime-agnostic since it's eBPF-based, but I see it's mentioned in the repo that it only supports:

> Java, Python, .NET, Node.js, and Go

Apart from Go (that is a WIP), these are the languages already supported with Otel's (non-eBPF-based) auto-instrumentation. Apart from a win on latency (which is nice, but could in theory be combated with sampling), why else go this route?

Otherwise known as the "bootlegger and baptist" problem.

Honestly someone should just fork the InfluxDB v1.x line and focus on stability and performance. The pre-1.0 to 1.0 release was painful, but what's there now is pretty solid as well as MIT licensed.

It blows my mind that they still have not learned their lessons with world breaking changes. How can you build anything on a database that is constantly reinventing itself chasing some new hotness?

I always thought the "PE" was for Placebo Effect which would explain why it never does anything.

The 4% GP quoted was in reference to individual cancers (since there is no "cancer" screening, each screening is specific to a type of cancer).

I'm having trouble finding numbers, but the comment by Vinay was basically saying that each screening is testing for something that has 1-4% chance of killing you. Not that 4% of all cause mortality is cancer, which is incorrect. In the context of individual screenings per cancer, the numbers roughly make sense based on what I could find, but I am by no means an expert.

The specific video/content they are rehashing is the one by Vinay here - https://www.youtube.com/watch?v=-9hQO7X1bmU

I'd highly recommend listening to the Econtalk conversation as there is a lot of nuance behind why screenings (at least as they are done today) could potentially be a net negative at the individual level and don't seem to have improved all-cause mortality in a significant way (according to Vinay).

I bought a house last month in Texas using a 30-year fixed loan.

We did use a RE agent, though I'm not convinced they are worth the ~6% they get on the transaction (3% for seller, 3% for buyer), especially on the buying side.

We put 20% down and then payed down to get to a ~5% rate. This was a gamble, but it looks like it'll pay off over the next five-ish years as the rate we would have gotten otherwise would have been substantially higher (as rates go). It was previously a rental home, so not in great shape - but not what I would call a "fixer-upper".

Sometimes you just need to decide whether the home, layout, neighborhood, school district, loan, etc is worth it and, if so, go for it. Assuming you are being responsible to your obligations (kids, for example), live your life the way you want to live it.

Just an FYI it might not need an app but the Amcrest cameras I have do phone home pretty aggressively. I black-holed them on my network, but no idea what kind of data it might be sending back.

I mostly agree, but I think your response misses the point that tons of people _believe_ that climate change is an existential threat - regardless of what the truth is.

I think a lot of people (myself included) would feel more at ease if we as a society were doing _anything_ to combat the projected worst case scenarios or at least give the illusion that we have things under control. It feels like every week there is a new story about some concerning weather event (fires, heat waves, droughts) giving more evidence to the doomer perspective.

I didn't know Zulip was acquired by Dropbox! Thanks for the info.

It's interesting to imagine a timeline where Dropbox becomes a leading business/collaboration tool, similar to GSuite or MS Office. It seems like they had many of the right pieces, between storage, paper, zulip... If they had added email and calendar plus potentially their own presentation and spreadsheet tools they could be have been a contender instead of relegated to pretty just storage at this point.

I'd recommend starting here:

https://writings.stephenwolfram.com/2023/02/what-is-chatgpt-...

It's pretty lengthy but doesn't require a PhD to understand. If you can get to the end of it you'll have a much better understanding of what's going on.

That's not cheating? That sounds like the interviewers could see his potential and guided/corrected him through the process. His willingness to take feedback and keep going in itself would be something to be commended, especially for a new grad.

What's discussed in the post is clear deception of the interviewers by the interviewee (using a second laptop to google answers), which is a very different thing.

The only part of overlapping functionality between DataDog and Otel is the agent.

In theory you could use the Otel Collector (or any other Otel-compatible agent) instead of the DD Agent to collect metrics/logs/traces. This would then make it easier for you to switch from DD to another Otel-compatible provider (Grafana, for example)... but 99.9% of what DD provides is _not_ the agent, it's dashboarding, alerting, RUM, synthetics, etc.

Basically Otel has made _agent_ switching costs effectively drop to zero, but that is a very small part of the whole picture. Like I said above, this primarily hurts vendors with proprietary agents that can't/won't adopt Otel for ingesting data.

Sentry is strictly code-first APM, which is only a part of what DD provides. What "APM" _is_ can get kind of blurry, but they are not direct competitors in a meaningful way.

OpenTelemetry (or OTel) is in no way an existential threat to DataDog. Primarily because OTel is simply the substrate/protocol by which data is collected from your apps/systems. DD does _a lot_ more than what OTel provides (RUM, SIEM, synthetics, on-call, dashboarding, anomaly detection, and much much more). If anything it's an existential threat to the more legacy vendors that aren't equipped to provide an OTel ingest layer.

OTel _does_ prevent lock-in on the agent side (making it easier to switch vendors) with open source components and consistent schemas, but OTel doesn't enable you to do anything that you couldn't do before with a specific vendor. It just empowers you to take ownership of your observability data, should you want to. Many don't, though. They want to throw money at someone else who can do it relatively well, hence the ridiculous DD pricing.

Regarding:

> Not sure how companies rationalize these types of services at scale when there are so many open source options to run for a fraction of the cost

At Coinbase's scale (which I assume is a lot due to the DD bill, but I haven't looked closely at it) the open source options simply won't cut it. Plus there are no scalable open source options for many of the things that DD does (synthetics, SIEM come to mind - not to mention onerous regulatory requirements). 65M seems like a lot, but it also means their cloud costs are insane - so maybe it lets them put focus elsewhere?

This points to needing _more_ regulation, not less. Potency could be mandated with clear warnings/notices at the point of purchase (like alcohol). The FDA (or whatever governing body) could also limit the potency.

The current status quo is that no one knows what they are getting until it's used, at which point its too late. If you could know exactly what you are getting (plus the means to seek help without fear of legal repercussions), then this would be a non-issue.

You could also have made a similar case with alcohol, but at some point the potency makes the product worse. For example, how many people do you see drinking everclear?

For anyone in the market for a remote, I have one of these[1] and it's pretty awesome. I'll have to check out the Lenovo as well.

[1] https://www.amazon.com/gp/product/B07WJGSXT8

My experience so far is very similar to yours. The technology is _really_ impressive (we have managed to transform electricity into knowledge!), but to say it is at the same level as the atom bomb seems a bit premature. My impression (or maybe my hope) is that your thinking is in line with the "silent majority" of people watching the hysteria from the sidelines.

My personal experience is that the GPTs is that they are a better Google. Why people seem to think that the models' "intelligence" will start scaling exponentially beyond where it is today (somehow _vastly_ exceeding the intelligence of the humans that created the model/training data itself, no less) is beyond me.

Will the models continue to improve? I suspect they will. Will it suddenly turn into a vengeful god and enslave/exterminate us all? That seems like a leap. I think we will need a true Hiroshima-style moment with AI in order to change public opinion that far.

I wonder if there is something deep inside the human psyche that endlessly looks for and, at some level, _craves_ existential crises like this. We look for danger everywhere and project our own fears and anxiety at whatever seems to fit the bill.

> Google only has a fraction of the training data.

Uh, what? The same Google that has been crawling, indexing, and letting people search the entire Internet for the last 25 years? They have owned DeepMind for nearly twice as long as OpenAI has been in existence!

If anything this is proof that no one at Google can get anything done anymore, and lack of training data ain't the problem.

I just wanted to say great job!

I would love to read a small write-up of how you put it together if/when you have the time. So far most of my experimentation with Stable Diffusion has been lackluster at best, though I haven't tried doing a composite yet.

> What is so bad about being aware of what content we show children?

That is not and has never been the problem. If you are a parent of a young child, then you should absolutely be aware of the content your child is seeing/reading. I doubt you will find anyone that disagrees.

The problem is the revisionist approach to rewriting works to appease modern readers' (apparently delicate) sensibilities.

Personally, I see this as shielding readers from harsh truths of the past (and in some cases just reality). Mr. Dahl is long dead, and yet we feel like we need to rewrite his words to ensure we don't accidentally offend someone. I mean, can you imagine a child reading the word <looks around> "fat"? Or "female"? My god, they would never be the same.

This is called 'crony capitalism'.

Not to disparage the work here, but my advice to "geek dads" (or any parent) is to put effort into A) researching and B) buying (paper) books that contain stories that _you_ want your children to hear.

Humans are creatures of story, and bedtime is one of the few precious moments in your child's day where you can have their entire attention in a quiet and comfortable setting. Read them something you want them to hear, ask them questions about it, and listen to their thoughts. I guarantee that you won't regret it.

For anyone interested in where to start, I highly recommend anything by Arnold Lobel, Else Holmelund Minarik, Peggy Parish, Judi Barrett, and similar authors. You'll be able to find these authors at any secondary book store for cheap or buying used on Amazon.

Can you explain in more detail how Lightrun works? It sounds like monkey-patching in real-time from what I can gather, which is neat but surely comes with some kind of overhead.

> Then I asked it for some problems for the students to solve and it wrote those too.

And your students will use it to solve the problems! In the end no one will learn anything and nothing of value was created. Rinse and repeat.

I guess this is my concern - if everyone starts using LLMs for stuff like this, then what's the point of anything?

What does the response time look like for those queries? And what is the cost?

I haven't heard great things about Athena, but I'd be curious to hear more about what works.

No. I don't know how hckrnews picks and chooses what it shows, but it doesn't show _all_ posts. If you look at new in HN you'll see pretty quickly that there are too many posts to show all of them.

> All pumped storage requires is a big high spot and a big low spot

I think that's what the GP meant by "extremely geographically dependent." Finding a 1,000 ft height difference (using the Gordon Butte Project as an example) between reservoirs will be really challenging in huge swaths of the United States and elsewhere.

That sounds nice, but it simply is not possible for very technical products (many kinds of databases, low-level infra, dev tools, etc). You won’t be able to find BDRs that can talk to devs the way another dev would.

It varies drastically by the business but I would say that it’s more on the magnitude of 10% or less (for enterprise sales, IME).

If you’re getting a 50% conversion on your demos then either your sales org has _really_ dialed in the target persona (qualifying everyone else out early) or your market is very wide.

It looks like these were made with https://excalidraw.com/

Highly recommended!