hbrn

29 ms·

301.
by hbrn 4y ago
> Also, filenames and directories tree act as metadata to help create a mental map of the application Correct. But this is an argument against splitting: once your folder structure reflects your mental model, you should no longer split,
302.
by hbrn 4y ago
> neither Carmack nor Blow are notable authorities on systems that have to be maintained by changing teams of hundreds of people that come and go Most systems don't have to be maintained by hundreds of people. And yet they are: mayb
303.
by hbrn 4y ago
The right answer is that there is no right answer. You shouldn't divide your code based on arbitrary metric like size, you should divide it based on concepts/domains. If a particular domain gets big enough it probably means it c
304.
by hbrn 4y ago
I agree, but the problem is that eventually it becomes not okay. So it requires a bit of nuance to understand when it is and when it isn't. Unfortunately most engineers don't like nuance, they want one-size-fits-all solutions.
305.
by hbrn 4y ago
> That problem? Not using statically typed languages. TDD was invented by a Java programmer. How does that fit into your world view?
306.
by hbrn 4y ago
> What happens if you need to redesign the architecture to meet new needs? How often does major reorganization happen? For most companies the answer is never. The "what if" argument is what leads to all sorts of premature optim
307.
by hbrn 4y ago
A lot of complexity is coming from big corporations producing frameworks and methodologies optimized for huge (bloated) teams / codebases. The reality is that most teams in the world are not huge and never will be.
308.
by hbrn 4y ago
A slight caricature of your take: It took us 20 years and immense complexity to start displaying loading indicators on buttons.
309.
by hbrn 4y ago
What's interesting is why was it so widely accepted in the first place. It's not like boilerplate reveals itself at later stages, it's right there from the beginning. I think it tells a lot about the state of the industry.
310.
by hbrn 4y ago
> Almost no other engineering discipline is volatile like that. So why should SE be? Because it can afford to be like that. You seem to be treating volatility as something negative. But it's only a side effect of software engineerin
311.
by hbrn 4y ago
> The biggest obstacle is not grokking that programming is a formal system, just like math I think there's danger in this mindset. Unlike math, programming is always directly tied to real-world problems. And those are defined by hum
312.
by hbrn 4y ago
I think the reality is that programming is not as intellectually challenging as we'd like to think. Maybe we will encounter a hard problem occasionally if we're lucky, but most of them are already solved. As programmers, we just n
313.
by hbrn 4y ago
It doesn't seem to be that hard, but keep in mind that staking requires you to be actively validating. There are major penalties for screwing up (attacking the network), and minor penalties even for going offline. You still need hardwa
314.
by hbrn 4y ago
> Apple doesn’t have the patent on this. But they do: D558,757 and D670,286. Don't people remember Apple v. Samsung case? https://patents.google.com/patent/USD558757S1/en https://patents.google.
315.
by hbrn 4y ago
This is very true. I've been in the same position and eventually I just burned out. If you think a programmer can become a manager, you can guide them, but can't force them to become your vision of a manager. The reason they can
316.
by hbrn 4y ago
There's an inherent gap between managers and programmers, which does indeed makes programmers the best managers. As a manager, you don't know what does "better architecture" mean. You don't know which practices of y
317.
by hbrn 4y ago
These days are long gone. Most programmers are building products for people.
318.
by hbrn 4y ago
Again, I agree they are incapable of creating products. I just think that entry barrier for AI is so much higher that Google has a decent chance. Amazon or Airbnb could be started in a garage by a small group of enthusiasts. Something like
319.
by hbrn 4y ago
> Google lacks competence to create any product that’s not adTech. While I wholeheartedly agree with this statement, unfortunately AI is an area where they might succeed. AI mainly requires: 1. Brilliant researchers/engineers 2. A t
320.
by hbrn 4y ago
Correct. I doubt they can just give it to you at face value. Are there other employees that were incentivized with stock/options? Were there investment rounds? Were you with the company from the beginning? As a CTO, you can be entitled
321.
by hbrn 4y ago
> it may be more impactful to make an app designed for them, which then plugs into the independant smaller networks of drivers seamlessly My understanding is that Uber already does that, though I don't know how big is the scale. The
322.
by hbrn 4y ago
> Doesn't "CSO" come with potential for personal liability in worst-case scenarios? Depends on your definition of worse-case scenario, but generally no. Otherwise no one would be dumb enough to take that job. Personal liab
323.
by hbrn 4y ago
Agree with everything except the last bit. As an early employee you have way more influence on company success than investor. You also have way more visibility into it if you decide to jump ship. Generally speaking, in terms of risk early e
324.
by hbrn 4y ago
> I'd rather have shares At this stage you're not going to get shares, unless they really want you. What you should be getting is options (with strike price at current/last valuation). Think about how likely is the exit
325.
by hbrn 4y ago
This. Security is one of those things everybody thinks you can't have too much of. But of course you can. You can literally kill your company if you start blindly implementing "best practices". It's very easy to fall int
326.
by hbrn 4y ago
It is, but Uber & Lyft got a billions of cheap dollars they spent on growth. Your growth is going to be way slower than theirs. There's also chicken & egg problem: no-one is going to use your app if it only has 5 drivers, and o
327.
by hbrn 4y ago
> would it also be able to scale with increase in visitor traffic? would it be able to deal with increase in media uploads Yes. It's not a social network / advertising machine. Hosting is a tiny fraction of Wikimedia spending:
328.
by hbrn 4y ago
> If you have any untested code and need to make changes that affect it, you are incredibly screwed But you aren't. If your software isn't landing airplanes, you shouldn't treat all bugs as critical. Build safety nets, not
329.
by hbrn 4y ago
> bunch of extremely well-paid SWEs complain about donating two bucks to a completely free encyclopedia You are NOT donating to free encyclopedia. You are donating to their parent company, which spends fraction of your donations on the e
330.
by hbrn 4y ago
Just to clarify, there's no Wikipedia Foundation, there's Wikimedia Foundation. Wikipedia is owned by Wikimedia Foundation. You can't donate to Wikipedia, the banners on Wikipedia lead to Wikimedia donations. It is quite unet

More ›