6,104 karma · joined September 8, 2019
One request from me would be for them to bring back the + operator and favour it instead of quotes, since Google Plus integration doesn't seem to be a thing anymore.
Your point is that "stuff <10% of people will like" doesn't make up for a mediocre film, and studios are trying to capitalise on that rather than working on better stories.
If so, I definitely agree.
I seriously hold no dog in this race but you're literally calling me a radical... maybe you're confusing me with someone else?
That sounds ludicrous. A bigger team or more money are incredibly difficult, especially for hobbyist and open source.
And the whole goal of this idea is to reduce the scope of projects in the first place!
It's impossible to reduce the scope if it requires something that might be as big as the engine itself.
-
> Lapsus linguae. I meant to write debugging. So read it as, debugging JS in JS VM, isn't as good as debugging JS in JVM and vice versa (JVM in JS).
But the Javascript doesn't run inside the JVM or inside the Electron process. It happens in a completely separate process! It's the same for the syntax highlighting: it happens inside the Language Server for VSCode, and it's probably the same for JetBrains when using Typescript too.
-
> Implying your example isn't cherry-picked? Really? How much have you used VSCode for languages with other VMs?
Of course it's cherry-picked, but this is how counterexamples work. The point is that a JetBrains IDE isn't by definition "better integrated", there are several more factors.
The previous CEO of the story wasn't wrong about what's important at all. Only about how much time it takes. He considered everything important and maximum priority, but didn't believe the "previous thing" was taking 80% of the time of the team. I couldn't tell the team to "drop the previous task" like the other post suggested: it was also maximum priority.
"Important" and "Takes too much time" are very different.
The thing they wanted me to do was maximum priority, but there was also some other totally unpredictable thing (often support related) that I also had to take care of. The higher-up was 100% sure it wasn't a real problem and assumed I was exaggerating the situation.
The solution to the problem of lack of trust is of course to convince them, or to leave. If you can't make changes that increase your productivity, you'll become stressed and burned out, and will still not be able to do enough. It is a decision they made.
In fact I know plenty of "woke" people who also think that it sucks. Some maybe semi-secretly.
The film has plenty to be criticised.
Sure, media is a huge problem. But people behaving like animals, having no empathy, taking no taking time to reflect upon what they say and taking no time to check other perspectives is what's doing the real damage here.
In general I don't blame Russians for what Putin is doing, but I will definitely blame a Russian that unapologetically supports the invasion of Ukraine, despite having the means to know otherwise.
If I tried to give them another perspective and they tell me to go fuck myself, or if they were actively trying to push an agenda (like it's common in USA political discussions), I would definitely blame them.
The problem is how the media (including social media) amplifies it.
Most people don't give a shit about it. Assuming the rest does is heavy projection.
Basically they store 100x more data than they need because of data duplication between microservices. The company didn't even have that much customers (<1000), data or even enough users (a handful per company), but duplication was still generating several terabytes.
It's amazing how much complexity people are able to cram into simple system.
(Interestingly enough there's another recent comment on another article here talking about this exact issue: https://news.ycombinator.com/item?id=32320660 )
About the other two: like you said, both React and Vue can still run old code, almost unmodified. The track record of those two is way better than most backend frameworks, by miles. It’s not even a fair fight.
Sure there's new concepts, but it's not arbitrary stuff or old patterns, it's legitimate stuff that makes writing UIs better. But the old ways are still there too.
And I'm with business people here: coordination between multiple devs is definitely an overhead that impacts businesses and heavily limits possibilities of experimenting and iterating with ideas.
The separation is also often a kick the teeth of Agile ("we gotta do the backend before frontend starts") and performance ("just call those 5 endpoints N times, we don't have time to make an aggregated endpoint, should have told us before"). You can still have specialists for micro-optimisations and fine-tuning that SQL query, but the initial implementation phase really suffers from the separation.
There's also lots of "prejudice" and pedantry around the term "full-stack", and that's almost always from less experienced developers or students (which is why you don't see it here in HN but see a lot on Reddit), but that's another discussion.
It's even worse with developers who had some design training, IME. "Oh you say you had design training but I'm 100% sure you're absolute garbage in either that or in database modelling. Or perhaps in both!"
I didn't know this story, so I searched for a video: https://www.youtube.com/watch?v=K6O_Ep9bY0U
Investors and CEOs think they need massive numbers because they misunderstand how dev headcount scales and how much overhead it generates. Hiring warehouse workers, moderators and content editors scale, so they extrapolate that to devs and expect the same results.
Also with lots of devs you need lots of designers, managers, product owners and other idea people, so there’s a lot of minor tasks being pushed trough, which would otherwise not be if it was a 20 dev product. Not to mention the extra need of support, QA and ops people, which in turn needs more coordination and management.
Going back to 10 or 20 devs is borderline impossible, even if you reduce the scope, since the infrastructure needed to support 100 or 1000 has too much of an overhead in itself. I would love to see someone try, though.
It is also cool to have 1000 devs in case of an acqui-hire. If your product sucks, you can still sell your team. Yay.
Be careful with that misquoting there: I said "different skillsets", I didn't say necessarily roles. An Ops and a DevOps person will know different things.
> Don't you feel that if you can hire a “DevOps engineer” for a specific role on a software team then we have all lost at using the phrase DevOps?
So, you copy-pasted this from the article. If so I'll assume you're agreeing with them and this conversation can be over.
However, even if we go by your definition that it is the same thing, then the "Dev" part is completely redundant and we should just call it "Ops", since it's by definition the same thing and it doesn't warrant a new title, because it's the exact same thing. However I doubt that's what the case the original article is talking about.
Moreover, in a lot of places there are "DevOps" teams that consist only of Operations people, which is even worse. The usage of DevOps is to attract candidates, not to differentiate the skillset needed.
Devops does involve operations, but is not exclusively that.
Having an Operations person work together with developers doesn't automatically make the Ops engineer into a DevOps engineer. Just as it wouldn't make the non-operations developers into DevOps engineers.
Maybe the better term would be "Operations Engineer for multi-disciplinary Team". Not DevOps.
EDIT: It was previously written "It seems we both agree that having a Ops engineer is not incompatible with DevOps?"
I think you missed the crux of the issue: in a lot of companies, Agile is definitely waterfall-ish, with a lot of pre-planning, design working a few sprints in advance, a lot of "getting it perfect" during PR reviews, no follow-up, no evaluating the work midway, no admitting failure and reverting it before finished, etc.
> Having roles for operations is not incompatible with DevOps at all.
I think you also missed the problem on this one. If you have "roles for operations" that doesn't involve development, it's not a devops role, it's just an operations role. Sure you can have devops in the mix, but someone doing operations exclusively isn't devops.
There is a standard where you can use some keys in your TV remote to control your device via HDMI, but I never had a device compatible with it.
I guess developers who do it got the habit from their layperson days.