marcinzm

In my experience it's not just the judgement of the manager but the judgement of their whole reporting chain. All of which can read the reviews and may be the ultimate decision makers on promotions/ratings.

My cynical view is that's it's to find scapegoats especially in companies that have a lot of politics or a lot of PIPs.

You generally scapegoat one level down and then let that level push it further down if it can.

So the Directors need team level metrics to find which managers reporting to them to scapegoat and have data to "prove" it.

Then the Managers need individual level metrics to find which engineers to scapegoat and have data to "prove" it.

This is my understanding of the US situation.

It used to be that doctors were running their own small medical offices a lot. Unfortunately, due to insurance overhead it's no longer profitable for individual doctors to do that, especially without existing patients, so the remaining offices are mostly from doctors near retirement age.

So as doctors retire and new doctors don't want to start their own practices those practices have been getting bought out by large medical networks or private equity. Some offices are also going out of network with insurance companies. It doesn't help that PE is willing to pay a ton more than any new doctor could for the practice.

I'd recommend picking medical plans with out referral requirements and with good out-of-network coverage.

> Presumably, once the copyright is allowed to expire, it can't be sold and then reinstated.

Derivative works have their own copyrights. The original book is free for grabs but each movie franchise has it's own copyrights. Likely far more valuable and ones the original artist will never see a cent from.

That seems to benefit large corporations at the expense of smaller artists. Either you focus on making money or some large corporation will swoop in the second you can't and exploit your work for their own profit.

> I think in this sentence, there's a hidden assumption that most projects look like your project(s). That's likely false.

You left out the part of that quote where I explained my assumption very clearly: A decently designed system.

In my experience if you cannot split tasks into <1 week the vast majority of the time then your code has massive land mines in it. The design may be too inter-connected, too many assumptions baked too deeply, not enough tests, or various other issues. You should address those landmines before you step on them rather than perpetually trying to walk around them. Then splitting projects down becomes much much easier.

> In any case, the nature of the product / features / refactorings usually dictates the minimum size of a PR.

Why not split the big tickets into smaller tickets which are delivered individually? There's cases where you literally can't but in my experience those are the minority or at least should be assuming a decently designed system.

That's also been my experience. It's part time work for a single on call engineer on a team of 6-8. If it's their full time work for a given sprint then we have an urgent retro item to discuss around bug rates, code quality and so on.

It feels like having 50% of your team's time be spent on urgent support, triage and bugs is a lot. That seems like a much better thing to solve versus trying to work around the issue by splitting the team. Probably having those people fix bugs while a 4 week re-factor in a secluded branch is constantly in process doesn't help with efficiency or bug rate.

It's I think sort of clear from everything that Automattic's leadership doesn't seem to understand that running a business means you're running a business. You're not making a side project, working on a hobby or running a non-profit. You're running a for-profit business which means others will treat you as such and you can't make excuses for it.

As I see it, aside from the obvious (Security, food, etc.) money lets you focus more on certain activities of your own choice. Either through more time or more efficient use of time (ie: pay for a coach, etc.).

This can make you happier if those activities align with what brings you happiness in a sustainable way. Many people are pretty terrible at knowing what that is or caring to do those activities which just makes it a net neutral. Cocaine fueled benders instead of a top tier therapist type of stuff.

Anecdotes are not data, and requiring perfection is a really odd bar for working or not working.

Tensorflow works really well in theory. In practice a lot less so. I saw someone spend months fighting Tensorflow to convert a production model from CPU to GPU inference with any sort of efficiency. Tons of issues due to bugs across versions, deprecations of features across versions, the graph optimizer shuffling data back to the CPU for no decent reason, etc. The person had no idea what was happening or why most of the time due to how black box Tensorflow was. This was a very senior ML engineer with a lot of Tensorflow experience.

Cursor IDE with Claude 3.5 has been very beneficial for me in terms of productivity. Others a lot less so.

This is what the repo says:

>Results >Ariane RISC-V CPU >View the full details of the Ariane experiment on our details page. With this code we are able to get comparable or better results training from scratch as fine-tuning a pre-trained model.

The paper includes a graph showing that it takes longer for Ariane to train without pre-training however the results in the end are the same.

May have something to do with who the supplier of natural gas to Europe used to be.

As I see it, Rich's stance is that of an expert in the database that doesn't need to deliver business features using the database. New users are not experts and even experienced users that work for companies have pressure to deliver features. You can get initial popularity by targeting these types of expert users working on more experimental products. However long term growth and popularity requires targeting the other 99.9% of users. I've seen one company adopt Datomic due to this type of user and then a couple years latter rip it out because as it grew it's developers were no longer of this type.

Isn't that the case for almost every single VC backed startup?

That proves the opposite point. At small scale FAANG solutions are non-sensical. At 100 petabytes good luck running your code on a laptop.

> Wild, I would literally never work for a company where I had this kind of relationship with my direct manager. My direct manager should be in my corner and on my side always. They will get me the highest pay, highest title, best perks, lowest amount of stress and work, most time off, best benefits they can and anything they ask me to do is in service of that. If I find out that my direct manager is the one gating a raise or promotion from me I'll just quit.

Nothing is infinite and managers are both people and have their own constraints. In my experience, expecting otherwise will just select for either managers who work really hard but fail to get you much (due to lack of political skills), or those that are very good liars that say they're helping you but never do (and you never realize otherwise).

Isn't that currently the case except for a very small number of people?

Is that so different than now for all but a few human elites?

> Generalists often innovate better than specialists

So hire generalist seniors? They exist.

Sounds like the author wants to patch a broken hiring system (only specialists get hired as seniors) by throwing juniors at it. Throwing juniors is a time-honored solution to many problems but while it tends to be somewhat viable and quick to implement it's also a sub-optimal solution.

Fortune 500 executives aren't the idiots you think they are. Some are but most aren't and most are moving to cloud.

> What large corporation would ever sign up to that?

Ones with the power to have some very nice contract terms put into place to mitigate downsides in the medium term. If it's a problem long term then that'll be the next CEOs problem and the current one will have cashed in their $50m check already.

Less cynically, for many of them it's either being captive to their barely working IT department or being captive to a decently more competent AWS. So they choose the one that will give them better features for less money in the medium term. They can't make their IT work better because it's a two sided monopoly. Only one client who is guaranteed to not leave (until AWS at least). That inherently and unavoidably creates horrible incentives.

The problem is that 90% of founders will take it as such right away versus reflecting on their own personal skill set first given how it's being framed.

As I see it, certain founders can function well in founder mode due to their knowledge, personality and luck in term of business positioning. Most can't. Most founders also won't make $100b companies. The latter type of founder trying to work in founder mode will make things much much worse and often result in going from a $1b exit to $0b exit.

> doesn't make political decisions.

Twitter is made of people and owned by people who use it to make political decisions. That was the case before Musk and is still the case after Musk.

Between Elon deciding who to block (and he blocks quiet often) and democratically elected governments deciding, I prefer the one that's got at least some checks and balances.

That's for the nations courts to decide and not Elon. It's just an excuse to support authoritarian governments that have no checks and balances in their constitutions at all.