HN user
See the history of Reddit. It was manually curated sock puppets before bots were viable, and now that bots are viable, I strongly believe the bulk of comments and posts in the popular subreddits are bots (and many are run by reddit themselves).
I've recently used a ruby script with a bunch of methods that I can configure with a few yaml files to generate a shell script which I can then run.
Each run generates a timestamped folder with not only the bash script, but also a copy of the configs used to generate it, and all the data that the commands in the bash script needs (json files). I find this style of generator a common pattern for commonly used but frequently tweaked scripts.
Location: SF Bay Area
Remote: Yes (willing to come to office occasionally)
Willing to relocate: no
Technologies: Java, Scala, Ruby, Python, C++, SQL, No-SQL, GraphQL etc
Resume: https://srikanthak.name/files/srikanth_resume.pdf
Email and other contact info: see resume
Worked on large cross-org projects at bit companies and built stuff from ground up at early stage startups.
Not interested in working on AI/ML or FinTech.
So you're paying them to not fix a bug in their system, am I reading that right?
The contract of the web is:
- I ask for a resource - you give it to me - any linked resources (stylesheets, scripts, images etc) are up to me to request
Therefore there is no "ethical" conundrum in blocking ads. The ad industry brought this on themselves by trying to push malware, spam and actively trying to make the web worse.
I made a simple one myself as a one file html app. https://gitlab.com/aksrikanth/flash-cards
Think of what happens to a fluid when exposed to a vacuum.
The water surrounding the vacuum instantly vaporizes and creates bubbles of water vapor. Then these small vacuum bubbles collapse and create small shockwaves that create noise and damage the propeller/screw.
Is there a firefox extension that can give me a GPT-3 summary of any given page? That would be super useful.
I'm now wondering if there is a time in history when this would not be seen as a coincidence?
Like a time when there was no crazy ruler.
Uber and Lyft are betting that 7/8 threshold will never be hit organically, but they can buy a 7/8 share of the legislature ensuring that this measure can only develop in one direction.
CSS is declarative and not imperative. A lot of people have issues with a different kind of programming.
Also the "cascading" nature of CSS makes it hard to reason about.
I love my Ricoh GR and bought it nearly 6 years ago. The camera is an amazing combination of an APS-C sized sensor in a pocketable form factor combined with a fixed focal length lens (which is superb) that is perfect for an everyday carry.
They only just came out with the GR III (the II was almost no change and didn't do well).
Their design choices are baffling. You could set up the GR to have tactile physical controls for all the things you may want to change on the fly, but the GR III has a touchscreen instead. It also has a 30% lower battery life (which wasn't exceptional to begin with).
Though I suppose the USB-C port is an improvement.
So any post that has "... a liberal amount..." or "... take the conservative approach ..." or "... replaced the right wing nut..." would be filtered.
Political candidates also share names with other people. Name based filtering can silence unrelated people. See how well our no fly lists are at targeting people who just happen to have the wrong name.
For every problem, there is a simple solution that is wrong. Getting things right is often not simple.
If you've ever used a firecracker rocket it has solid fuel in exactly the same way. It just burns from the inside out (I think) to ensure that the rocket's center of gravity doesn't change as the burn progresses.
Is it possible to have the hierarchical structure of tabs like in tree style tabs (so I can see which tab spawned which tabs)?
No one, not even those who love me, need to see my face in High Def.
It would be a good choice to replace tablets that are used outside conference rooms to show the schedule and/or book the room.
Also missing optimal placement of scarecrows for larger farms.
The first time I was a new (post) grad and I didn't much like it and left for a startup within a year.
I figured I'd give it another go with more experience, so the second time I was hired at L5 (expected to be leading teams). Liked it even less the second time, but stuck around for about 2 years for logistical reasons.
In my experience at Google and other large companies, the team you join ends up dictating your experience. I would be open to a third stint at Google, but would be very careful about vetting the team (especially the leadership).
Having said that, I don't think I like the culture at Google. Especially the second time, it felt very mercenary and there was cutthroat competition. It didn't feel like I was working on a cohesive vision, but just moving protobufs around. I think I preferred Google when Eric was CEO, maybe because it was smaller, but it felt like Larry was running Google with too much emphasis on the bottom line and not enough on product or vision.
This may have been true 15 years ago when they were a small company. Google now employs ~60K engineers and the majority of them are "Bog Standard".
Disclosure: I worked at Google twice.
Too little (PHP, JavaScript) is suboptimal, and too much (Scala, Haskell, Rust) is also suboptimal.
I find myself thinking the exact opposite. When you want to do something quick and dirty JavaScript or Ruby are my go to. When you want to do something right, do it in Haskell or Rust.
The middle of the road options I find are worse for anything.
Unless you want to hire big teams. Java, C#, Go are great for that.
“The reasonable man adapts himself to the world: the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man.”
― George Bernard Shaw, Man and Superman
I would really love if the system76 laptops had hardware kill switches for the camera and the microphone. If the camera switch also physically occludes the lens, that would be great.
Not necessarily true:
https://www.livescience.com/24694-identical-twins-not-identi...
paper (PDF) : https://www.researchgate.net/profile/Constantin_Polychronako...
Only on one of them:
- Your "friends" are known and tracked
- The content you view is determined by your browsing history, ad tracking and that of your "friends"
- The composition of the audience that's voting is poorly informed and susceptible to non-critical thinking (YMMV)
On both of them:
- Stories are posted to maximize your outrage/other emotional response
- Content of stories rarely matches the titles
Social interaction and even a like button is not the problem. What the like button does is the problem.
I use Firefox's Reader mode, and it fixes most of these "scroll all the time" sites.
Can we get a canonical url for this? It's a useful resource, but in its current form might disappear at any time.
Does this same logic apply to promote a Wealth Tax on people who have wealth that's being hoarded?
Funnily enough, when I do design interviews, I often ask them to design a chat client - not the whole design, but bits and pieces, like the db schema or the API of the service.
What the interviews invariably show is that a chat client is not simple. It's extremely complicated, especially when you want it to work at scale. If you think it's a "simple" app, go ahead, try to make one - even on a whiteboard.
I can completely understand why instant messaging can get you $700 million in revenue. As for the losses -- they are invested in the future of the company and I think are justified.
"How hard can it be?" is something we ask about "simple" stuff everyday, and the answer if you dig deeper is almost always "Much harder than you think."