Personally, I've had great gains in terms of small personal tools built on top of CLIs, and Emacs config. I'm also able to deliver PRs that demonstrate a general principle to someone I manage even if they work outside of my most comfortable stacks. But that's not what you asked about.
I don't have direct evidence of exactly what you're looking for (particularly the part about "someone responsible for the architecture to sign off"). Sticking strictly to that last caveat may prevent you from receiving some important signal.
> the claim that we should move from “validating architecture” to “validating behavior.”
I think these people are on the right track and the reason I think that is because of how I work with people right now.
I manage the work of ~10 developers pretty closely and am called on for advice to another ~10, while also juggling demanding stakeholders. For a while now, I've only been able to do spot checks on PRs myself. I don't consider that a major part of my job anymore. The management that is most valuable is:
1) Teaching developers about quality so that they start with better code, and give better reviews to each other
2) Teaching people to focus and move in small steps
3) Investing in guardrails
4) Metrics, e.g. it doesn't matter what code is merged, it doesn't matter if a "task" is "shipped", what matters is if the metrics say that we've had the result we expected.
As I acknowledge how flimsy my review process is, my impulse is to worry about architecture and security. But metrics and guardrails exist for those things too. Opinionated stacks help, for instance SQL injection opportunities look different enough from "normal" Rails to mean that there are linters that can catch many problems, and the linters are better than I am at this job.
Some of these tools are available for agents just as they are for humans. Some of them are woefully bad or lack good ergonomics for agents, but I wouldn't bet against them becoming better.
I agree that agentic coding changes code review, but I don't think that has to inevitably / long-term mean worse.
> half of my time went into fixing the subtle mistakes it made or the duplication it introduced
A cold hard evaluation of the effectiveness of agentic coding doesn't care about what percentage of time went into fixing bad code; it cares about the total time.
That said, I find that making an agent move in many small steps (just how I would advise a human) creates far less rework.
I can’t imagine any role in software that gets better delivering more work in longer cycles than less work in shorter cycles.
And I can’t speak for technical writing, but developers do operations and testing because automation makes it possible and better than living in a dysfunctional world of fire and forget coding.
I suspect that the policy is popular with the 90% of voters in the UK who earn less than £80k and that politicians are not very concerned with the ambitions of the rest of us (frustrating as that is when paying London rent).
I don't really think people understand it, it's quite hard to explain and on paper it just sounds like the person is wrong.
The maths for one child for me works out at ~£10k childcare bill without the free childcare vs ~£4000 with free childcare + tax free childcare (for three days a week). Even that doesn't sound too bad but it's the combination of that, the loss of personal allowance and the fact I still have a student loan that means the actual number of how much I have to earn to break even on losing it is something like £30k more than the actual cut off.
To me it appears as though the success of the right wing politics everywhere is that they made socioeconomically disadvantaged people identify other socioeconomically disadvantaged people and the middle-class as the cause of their suffering while somehow becoming sympathetic to the uber rich in hopes to one day belong. And to me it’s clear that if we taxed wealth and high incomes fairly and removed the loopholes to level the playing field we would not even need these discussions to begin with because we simply had a well financed social society and the rich would still be rich, but maybe not so obscenely so.
To be fair to the conversatives in the UK who have engineered this situation, some have recently said that the £100k threshold is too low. I detest them but I have to give them this.
We have 30 hours of free childcare in the UK (for nursery, schooling in older years is free) if both parents are working and neither earn more than £100k. It has the interesting impact that a salary of £99.9k is worth more to me than £130k, give or take some extra contributions to pension.
It’s interesting to me that the threshold is so much higher in San Francisco given that SF is only 8.7% more expensive than London, at least according to numbeo.
Maybe healthcare makes up some of that difference?
The £100k threshold is such an economically illiterate policy for society. The GPs and lawyers I know are working ~3 days a week to avoid it, so much economic output and taxes missed out on.
I am pessimistic that the reason it is so high is because someone making $220k per year said "yea but what about me, I have to pay for childcare too..." The number should be significantly lower. Anyone making a combined $200k a year has other options and opportunities, the immigrant family making $60k combined does not. This feels like a policy designed to "help" the poor but also benefit the rich..
A combined threshold of £200k (or whatever this converts to) would be great for me in London since my partner earns roughly the median salary (which is a lot lower).
Is there any truth to the climate sceptic claim that solar in the UK can’t generate useful power with our bad weather and cloud cover? I always said that it’s not like Germany has better weather but they have tons of solar. However it would be great to have an answer from the UK.
It definitely generates useful amounts of power, especially in the summer. The problem is that we still need gas plant, and the distribution infrastructure ready for the rest of the year when it isn't so effective.
Not that it changes your point, but the other day I met a republican who said he doesn’t think climate change is a thing but peak oil, now that’s something to worry about.
The longevity of this plus the “no anthropogenic climate change” nonsense is astounding. Armchair climate sceptics are happy to seriously stick to talking points that are so out of date that even the oil industry doesn’t use them anymore.
So you met a person, they told you their Party Affiliation, then went to tell you how it isn't "real".
Sorry I don't believe your paraphrasing of this person's real thoughts and ideas. I'm sure these people exist, but it doesn't mean anything. I could equally go find someone crazy saying the world is going to end this year.
We use 99% Apple and used to have a rental contract. We switched to just buying new MacBooks outright every 3 years because it was cheaper. I think the number of “it needs repair and I’m losing working hours” scenarios was very low.
It probably helps that everyone including (especially) the CEO can reinstall an OS and manage their own dotfiles.
I don't have direct evidence of exactly what you're looking for (particularly the part about "someone responsible for the architecture to sign off"). Sticking strictly to that last caveat may prevent you from receiving some important signal.
> the claim that we should move from “validating architecture” to “validating behavior.”
I think these people are on the right track and the reason I think that is because of how I work with people right now.
I manage the work of ~10 developers pretty closely and am called on for advice to another ~10, while also juggling demanding stakeholders. For a while now, I've only been able to do spot checks on PRs myself. I don't consider that a major part of my job anymore. The management that is most valuable is:
1) Teaching developers about quality so that they start with better code, and give better reviews to each other 2) Teaching people to focus and move in small steps 3) Investing in guardrails 4) Metrics, e.g. it doesn't matter what code is merged, it doesn't matter if a "task" is "shipped", what matters is if the metrics say that we've had the result we expected.
As I acknowledge how flimsy my review process is, my impulse is to worry about architecture and security. But metrics and guardrails exist for those things too. Opinionated stacks help, for instance SQL injection opportunities look different enough from "normal" Rails to mean that there are linters that can catch many problems, and the linters are better than I am at this job.
Some of these tools are available for agents just as they are for humans. Some of them are woefully bad or lack good ergonomics for agents, but I wouldn't bet against them becoming better.
I agree that agentic coding changes code review, but I don't think that has to inevitably / long-term mean worse.
> half of my time went into fixing the subtle mistakes it made or the duplication it introduced
A cold hard evaluation of the effectiveness of agentic coding doesn't care about what percentage of time went into fixing bad code; it cares about the total time.
That said, I find that making an agent move in many small steps (just how I would advise a human) creates far less rework.
reply