This may be a cynical take, but as someone with 10+ years of experience why should I care if companies are too short sighted to value and train juniors?
Wouldn't it be uncharitable to assume that the commenter is totally selfish and short-sighted? :p
It may be a cliche, but it's all connected. In a general sense, programmers at different experience levels are at least partially substitutable goods. A crash in wages on one group will probably affect that other.
In a more specific sense, companies won't pay seniors for skills at mentoring and managing the juniors they won't have.
Longer term this kind of stupidity will destroy the economy from both ends.
Unless there's an unexpected jump in AI IQ, vibe-coded projects will start to unravel, but the companies won't have the resources to hire the human coders needed to fix the code.
Meanwhile a lot of people with real skills and ability will have been unemployed long enough to depress spending across the entire economy.
Those same people would have been prime drivers of spending, because they were one of the few demographics with significant disposable income and the ability to afford high rents and property prices.
You can see where this is going.
The people running the companies can't. Or if they can, they maybe believe they have an escape route.
That will turn out to be a fantasy too.
The problem isn't AI. it's an economy running on fantasy numbers that are unmoored from economic and physical reality.
Because engineers enjoy tinkering more than anything, and they love telling everyone how fun tinkering is, and there was a narrative that tinkering was empowering and everyone should know how to do it, with a side order of "And if you get really good you can build a business and become super-rich too."
But the reality is law is primarily about social capital, medicine has more of that than most people realise, and computer people love to pretend social capital is something other people do, and they don't need to.
Can't really practice medical or legal skills from the comfort of your home on devices everyone has had for 10 years now. Also, unlike computers which must obey every whim of their master, lawyers and doctors deal with humans. Which makes raw skill less important.
Where do you draw the line on that attitude? Do you not care about global warming because in your lifetime, you're probably not going to experience an unsurvivable wet bulb temperature where you happen to live?
I draw the line at things that directly impact my net worth.
> Do you not care about global warming because you're probably not going to experience an unsurvivable wet bulb temperature where you happen to live in your lifetime?
Correct. I don’t care about global warming or climate change.
If I decide you're having a negative impact on my net worth, can I come to your home and shoot you in the head?
It seems we need a remedial class in morality here, where we work up to you understanding the golden rule. But perhaps you're not capable of understanding that. Is euthanizing you then the only option?
Nothing worse than a person who takes for granted all the hard work others put into society while smugly bragging about how they don't care about anyone but themselves.
It's a self-solving problem, though. At that point, every remaining senior+ engineer will be paid a bajillion dollars (like they are now) and companies will start to invest in actual training.
That's ... an interesting observation. I've found that LLMs work great when they can check their work and have a clear notion of what correct looks like. e.g. a good test suite they can rely on.
I'm sure a REPL helps out with this sort of thing quite a bit!
I fully agree that most people won't be comfortable if they're the "only" man or woman at the party.
How do you suggest evening genders out? Inviting couples is fine by definition; but I only really know single men. It seems odd for me to randomly invite single women I don't really know.
How would you handle the case where you want to trace 100% of errors? Presumably you don't know a trace is an error until after you've executed the thing and paid the price.
This is correct. It's a seemingly simple desire -- "always capture whenever there's a request with an error!" -- but the overhead needed to set that up gets complex. And then you start heading down the path of "well THESE business conditions are more important than THOSE business conditions!" and before you know it, you've got a nice little tower of sampling cards assembled. It's still worth it, just a hefty tax at times, and often the right solution is to just pay for more compute and data so that your engineers are spending less time on these meta-level concerns.
I wouldn't. "Trace contains an error" is a hideously bad criterion for sampling. If you have some storage subsystem where you always hedge/race reads to two replicas then cancel the request of the losing replica, then all of your traces will contain an error. It is a genuinely terrible feature.
Local logging of error conditions is the way to go. And I mean local, not to a central, indexed log search engine; that's also way too expensive.
I disagree that it's a bad criterion. The case you describe is what sounds difficult, treating one error as part of normal operations and another as not. That should be considered its own kind of error or other form of response, and sampling decisions could take that into consideration (or not).
Another reason against inflating sampling rates on errors is: for system stability you never want to do more stuff during errors than you would normally do. Doing something more expensive during an error can cause your whole system, or elements of it, to latch into an unplanned operating point where they only have the capacity to do the expensive error path, and all of the traffic is throwing errors because of the resource starvation.
You can use the OTel Collector for sampling decisions over tracing, it can also be used for reducing log cost before data is sent to Datadog. There's a whole category of telemetry pipeline now for fully managing that (full disclosure, I work for https://www.sawmills.ai which is a smart telemetry management platform)
It's more accurate to say those do the boilerplate of memory access necessary for complex types for you. You're still basically limited to integers and floats.
But when you think about it, isn't that basically true for native languages?
They're a way to assign ownership to individuals or teams on a granular basis, rather than at the repo-level. You can assign entire folders or individual files to people.
100% I've had great success inviting people to dinner parties and such. This works great for people I already know.
I've tried hosting public groups, with partial success. Have you got any ideas there?
I've tried posting events on local Facebook groups, mostly. But my experience is that very few people are actually active online and willing to physically do something.
Success really depends on the pool you're drawing from. A public open invite on FB is one of the hardest approaches. Kudos to the people who can pull it off.
The easiest for me was using my city's discord server as a launch pad for running book clubs, coffee meetups, and hikes. After a few successes, it's easy to gain a reputation for making things happen and it sort of snowballs from there.
It is also important to not become bitter when you fail. Starting up community activities is hard. It is somewhat like seeding servers in classic multiplayer games like CS1.6.
> I've tried hosting public groups, with partial success. Have you got any ideas there?
It is way easier if you give away free food. Like, it could be coffee or donuts. It just have to be some token value, such that people don't feel that they only go there for the nominal reason if the event sucks.
When my local party has like booths talking with voters we have like cookies and coffee such that there is an indirect reason to stand there talking to us. Some sort of plausible denialability.
Also, there is the automatic social aspect of food.
In areas where the website is still viable, I've found that Meetup is a reasonable way to socialize and find recreational events with limited pressure in a public location.
For example, I enjoy board gaming, and used to use Meetup in the South because I found it difficult to make friends being an outsider in the South. Meetup was relatively helpful for going to a restaurant, just sitting down, having meal, and playing board games for several hours. And several of those then turned into roleplaying groups, social groups, and similar external to the Meetup.
On a quick search, most major cities seem to have board gaming groups:
Most are relatively free of cost, supported by businesses that have an interest in regular attendees visiting the stores and buying a few small products each time they attend (meal, drinks, snacks, ect...) A couple are gaming stores that ask for a small rental fee or offer free table space if you want to bring your own games.
