greggman3

no profile record

コメント

4 年前議論

Name any apartment complex on a quiet street in SF that's not in Mission Bay that's been built in the last 15 years

4 年前議論

AirBnB sux but zoning has arguably been a net negative on most societies in many ways

https://youtu.be/bnKIVX968PQ

https://youtu.be/CCOdQsZa15o

for me what I noticed lately is that in California it's impossible to have a quiet modern apartment because nearly 100% of modern apartment buildings are only allowed to be built on busy noisy high traffic streets. If you want a quiet street you're forced to buy/rent a house

4 年前議論

They removed all of mine as well. Stayed around 24 places, half had issues, 3-4 were bad issues and left a bad review. AirBnb removed them

4 年前議論

It's not the "like" button. It's the "likes" number. Hide the number and the like button is signal to the system. So "likes" and it's a signal to the poster. That IMO is the problem, not the button itself, the "score" you see visually.

I've hidden the score here on HN for myself and on StackOverflow. Two places where score has a negative effect on myself.

4 年前議論

I agree with you. I ran into that issue once. Of course I knew map takes a function that it passes value, index, array, I just forgot for a moment. It would arguably be better to have different functions for those.

Very tangentially related, Apple's Metal API has a function that copies a texture. You pass it a width, height, etc... But, if the texture is compressed, then 255 of 256 possible value combos you pass it will be invalid since compressed textures can only be copied in block multiples. I think it would have been a better designed function if it only took width and height in blocks instead of pixels (with uncompressed textures defined has having 1x1 pixel blocks). Then this nonsense of 255 of 256 values being bad would disappear. There's a ton of other inconsistencies in that function. For example, you pass it destinationBytesPerRow when copying to a buffer but if the texture is compressed you pass it say 40 rows and it will only only actually copy 10 rows of blocks and only advance the destination every 4 rows instead of every row. It's arguably a poorly designed function. Thought, I suspect it was inspired by similarly poorly designed functions in other graphics APIs

4 年前議論

Different perspective, just go to the track and drive one. For ~$300-$500 it cured me of ever wanting to own one. I got to drive it hard in ways I'd likely rarely get to drive in if I owned one and it was so much cheaper.

5 年前議論

> 1. Tests should not be published

tests are published, they just are currently used by anyone but the owner

> 2. You can update the test script with the release too: "test": "echo pass"

which is why my suggestion was for npm to run the previous tests as well as the new tests. Given that people will more likely choose tested packages vs untested a package that just had "echo pass" wouldn't get much traction. And, given my suggestion is that npm run the previous tests, you couldn't just change the tests to "echo pass". Your only opportunity to do that is a major version but major versions are not upgraded automatically in any form so people would be more likely to catch and flag.

5 年前議論

I know this isn't a solution but .... I wondered if npm should run the tests before it allows a project to be published. In in particular it should run the previous tests. If the current package is 1.2.3 and you upload 1.3.0 the 1.2.3 tests should pass on 1.3.0 or else fail (not sure that's even possible to automate given the current design)

Even if work poor or no tests would get around this but hopefully people would chose well tested packages over poorly tested packages.

5 年前議論

Is it? In retail, AFAIK, if a store sells you a defective product, they are liable (or at least partly liable). It doesn't matter that some other manufacture made the product. The point being, responsibility is shared.

You're responsible for every dependency you add to your project and that includes all sub-dependencies. Your users will sue you for not doing your due diligence. You may turn around and try to sue your suppliers but that doesn't absolve you of your responsibility.

5 年前議論

I'm not saying there should be crunch but there is limited money (no idea for The Witcher). If you get funding for X amount of money you calculate your salaries plus overhead and you have a deadline. Miss the deadline then you renegotiate for more funding but at some point the funders will just pull the plug and your team are out of jobs.

It's unrealistic to believe there's no reason for deadlines. For a FAANG company, sure, they're making so much money they can -- mostly -- afford to do things without deadlines but few video game studios are at a level of funding that they can spend another 3, 6, 12, months.

The internet claims Cyberpunk 2077 had 500 to 5000 people working on it. I'm sure that 5000 number is not full time but it's not hard to believe the 500 number. Just paying 500 people $15 an hour, 40 hours a week $1.2 million a month, not counting rent/power/taxes etc... And of course some large percent of those people are probably getting closer or $30 per hour $60k yearly salary) to $60 per hour ($120k a year salary)

Witcher3 claims 150-250 employees so your delay, shipping October instead of May is easily $4 million needed and I suspect that number is low. $12 million if it took an extra year as you suggested and that's a low-ball estimate.

5 年前議論

Ugh! Those sites will do absolutely nothing to teach you how to make something like this. Thebookofshaders and shadertoy are about solving a puzzle "given only a single function who's input is the position of the pixel being draw, draw a pretty picture with math". That is not remotely related to the techniques used to make something like this paint program.

I don't have a suggestion for the paint program in particular. There are a few S.O. answers that cover painting linked here

https://webglfundamentals.org/webgl/lessons/webgl-qna.html

5 年前議論

Is there a market for good networking equipment? If Ubiquiti was it and it's gone, and reading this thread there are no good alternatives, then it sounds like there is an opportunity for a new company.

6 年前議論

I agree the LG feels bad. I don't agree that Apple doesn't have the talent to solve this and make 30% lighter laptops.

As for weight, carrying a 4.4lbs notebook to and from work and to meetings never bothered me. Carrying the same notebook for days on end as a digital nomad, especially in hot and humid weather did.

6 年前議論

not according to apple's home page.

    iPad Pro 12.9"   1.391 lbs
    Magic Keyboard   0.5 lbs
    Total            1.891 lbs

https://www.apple.com/shop/product/MLA22LL/A/magic-keyboard-...

Apple doesn't list the weight of an iPad Pro but wikipedia does

https://en.wikipedia.org/wiki/IPad_Pro#Fourth_generation#Mod...

6 年前議論

I was hopeful they'd weigh less. An iPad Pro 12.9 inches weighs 1.4lbs. The 13.3 inch MacBook Air M1 is 2.8lbs. I'm sure there are reasons but there are companies, LG for instance, that make 2 lbs 13.3 inch notebooks and 3 lbs 16 inch notebooks. No I don't want an LG but I was hopeful for a 16" Arm Macbook Pro that weight 3 lbs. Now I'm pretty confident it will be the same as an Intel Macbook Pro 16" at 4.4lbs (heavy enough my messenger bag leaves marks on my shoulder and gives me muscle pains if I have to lug it around all day) Somehow, getting it down to under 3.lbs removes that pain. Maybe because it includes the power supply which is also bigger.

To put another way. They could have taken an iPad Pro, expanded the screen to 13.3 inches, added a keyboard and put MacOS on it instead of iOS and it would be probably 1.8lbs. I don't know what the tradeoffs would be but I was excited by that possibility. It didn't happen though.

6 年前議論

MetroidVanias though actually make sense. The game play is similar to Metroid or Castlevania. So taking that same direction a roguelike should be similar to rogue. I've seen zero MetroidVanias where the similarities to Metroid and Castlevania aren't 100% clear.

6 年前議論

Except its not understood whatsoever. When I read roguelike or roguelite I understand that it should be something similar to "rogue". Since I'm not actually a fan of rogue I will therefore look no further at the game since it was clearly labeled as being similar to "rogue"

What other "~lite" or "~like" in the entire language as a similar property of being nothing like the "~" part?

6 年前議論

it's not remotely rogue "lite" either. In the context of all other "lite" things a rogue"lite" would be a rougelike with less features, not a some completely unrelated game genre.

6 年前議論

I'm disappointed that game called "roguelike" is absolutely nothing like "rogue"

Might as well say "Schindler's List" is an "Indylike" or a donut is a "Pizzalike"