1,266 karma · joined December 27, 2017
You don't even have to be as organised as in the example, LLMs are pretty good at making something out of ramblings.
I agree it's best if working in isolation, but if you need to synchronise then estimations make sense.
If you need 3 months to implement something, and another team 1 week, and both need to be ready at the same time; then if you actually know those estimations the second team can wait until then and do something immediately useful in between.
That's just too slow for business in my experience though. Rightly or wrongly, they want it now, not in a couple of sprints.
So what we do is we put both the investigation and the implementation in the same sprint, use the top of the range for the implementation, and re-evaluate things mid-sprint once the investigation is done. Of course this messes up predictability and agile people don't like it, but they don't have better ideas either on how to handle it.
Not sure if we're not enough agile or too agile for scrum.
Not at all, it has one of the lowest rate in Europe along with the UK. It's very hard to get the building permit required to install one. Portable AC has had a boom those past few years though (because it doesn't require a permit).
I tried various combos of weight* reps, and in the end the optimum was somewhere in the middle because no matter how light the weight there was a limit for me at about ~150 reps.
In my case, the curve would be: total volume increases quickly initially at you go from max weight/1 rep to something like 20/30 reps, then something of a plateau as things equalise, then it goes down again as you reach the max reps threshold.
With good naming it should be pretty obvious it's a Foo, and then either you know the type by heart, or will need to look up the definition anyway.
With standard containers, you can have the assumption that everyone knows the type, at least high level. So knowing whether it's a list, a vector, a stack, a map or a multimap, ... is pretty useful and avoid a lookup.
And if you add the value to the customers, then it's through the roof.
Maybe the reverse is more clear: if air travel didn't exist, it would have a huge economic impact; a clear proof of the value creation. Airlines just happen to capture essentially 0% of it.
We're having troubles with memory usage when using SQLite in-memory DBs with "a lot" of inserts and deletes. Like maybe inserting up to a 100k rows in 5 minutes, deleting them all after 5 minutes, and doing this for days on end. We see memory usage slowly creeping up over hours/days when doing that.
Any settings that would help with that? It's particularly bad on macOS, we've had instances where we reached 1GB of memory usage according to Activity Monitor after a week or so.
There's probably generally a bit of correlation. But just because someone can be very focused and go to extreme lengths in one aspect of their life doesn't mean they can consistently do it in every aspect of their life.
My experience is that the vast majority of people do 0 research (AI assisted or not) before asking questions online. Questions that could have usually been answered in a few seconds if they had tried.
If someone preface a question by saying they've done their research but would like validation, then yes it's in incredibly poor taste.
If all the examples you can conjure are decades old*, is it any wonder that people don't really take it seriously? Software power the whole world, and yet the example of critical failure we constantly hear about is close to half a century old?
I think the more insidious thing is all the "minor" pains being inflicted by software bugs, that when summed up reach crazy level of harm. It's just diluted so less striking. But even then, it's hard to say if the alternative of not using software would have been better overall.
* maybe they've added Boeing 737 Max to the list now?
Ultimately, what we care about, from a grid policy perspective, is the cost to provide 99% (or whatever) guaranteed 24/7/365 power. Each energy source will have its own challenges in order to do that, and for solar availablity it clearly one of them. And yes, externalities should be factored in.
> Or can we just assume that for now we have a mix of sources where different sources have different pros and cons.
Of course, but the question is: what is the right mix for the right place. And saying "solar + storage" is cheaper than gas means very different things if it can only guarantee 60% availability like in England, or 95+% in the sunniest regions of the world.
The figures reported here come from (among others) Lazard' LCOE analysis. (https://www.lazard.com/media/xemfey0k/lazards-lcoeplus-june-...) If you have a look at how it's calculated, "solar + storage" only has enough storage for 4h for instance. You cannot really meaningfully compare it directly with nuclear which is much more reliable in itself.
What LCOE says is that the system will produce X amount of energy at cost of Y. It says nothing about when that energy will be produced. An energy source that produces 365 MWh on January 1st only, and another that produces 1 MWh every day, for the same cost, will have the same LCOE. The latter is, provided you can scale it, much more useful in practice.
Look, I'm not saying that solar is bad or we shouldn't do it. It's just that the "solar is cheap" thing which is regularly reported is a bit misleading. We've heard it for years now, and yet electricity prices around the world are mostly increasing. Clearly there's a mismatch, but where does it come from? And I think part of the reason is that the "ancillary" costs of solar have been underestimated. Sure, the energy straight out of the panel is very cheap, but if you need 10s of billions in grid upgrades and storage/backup to make it work in practice, then it should definitely be included in the comparison! Just like the externalities of fossil fuels should be.
"Degree of coverage of motorised road traffic infrastructure costs: 111%"
And it's not even safety critical code.
I haven't found it to be really useful so far, but it's also very little added work, so for now I keep on using it. If it saves my ass even just once, it will probably be worth it overall.
So lots of allocations/deallocations. If you're only storing a few key/value pairs long term, you won't have any issues.
But as far as I can tell, it's more an OS issue than really a Sqlite issue, simply doing malloc/free in a loop results in a similar behaviour. And Sqlite does a lot of allocations. We see a similar problem on Windows, but not as pronounced, and there we can force the OS to reclaim memory.
It's probably solvable by using a custom allocator, but at this point it's no longer plug and play the way the GP meant.
To give some example: Switzerland is roughly electricity neutral over a year, but there's a significant winter/summer imbalance of about 5TWh. To add enough storage to compensate this imbalance, you would need to:
- cover about 2% of the country in batteries
- build about a thousand pump storage stations: despite the Alps covering about 40% of the country, it's not clear if you would have enough valleys to flood
- hydrogen looks a bit more reasonable, if we don't look at the costs, you only need to store a few millions m3 of liquid hydrogen. The gas storage in Germany for instance are quite a bit larger than that, but hydrogen is also significantly harder to store.
And all of this is to use once a year essentially! None of it looks practical or affordable (a pump storage station costs a few billions a piece for instance).
So if you get a DB from somewhere else and want to open it in whatever tool you use for visualisation, all the queries/views/... will not work because they don't know about all those functions.
Anything that would really love free energy also cost a lot to build and maintain/operate besides electricity. So much that a few hundred hours of free (or even better than free) energy a year is far from enough when you need >90% uptime to make sense. Maybe it makes you go from 95 to 85%, but still clearly it's far more than there are sunshine hours.