debaserab2

Any time I look at the new features in Chrome, I ask myself, "whose actually using this stuff?". Even though Firefox usually keeps pace pretty well it's pretty common for implementations to differ, and thus you're on the hook for browser quirk issues in any new feature you adopt immediately. Not to mention I've always had the anchor that is IE to temper any hastiness to use a feature (even if you simply don't implement the feature for IE, that's still an extra error state to handle unless you're willing to drop IE support altogether).

It can't even be 1% of websites out there embracing this stuff. I'm guessing the biggest consumer for most of the new features in Chrome is probably Google web developers. There's quite a lot of things that only end up half implemented, too.

Although the RFC these days seems to just be a design spec doc for Google, I suspect you could carve it down to less than half of what it is and render 99.9% of websites out there just fine.

> Why would I? Amazon is going to be better at it than I am.

Until it's not. Then suddenly you're trying to decipher cryptic cloud provider error messages in a service that made a false promise to you that it's abstraction was so air-tight that you'd never have to learn the underlying technology at all.

Then suddenly, you do need to know the underlying implementation, and quickly.

I use a chrome plugin[1] to ensure that I always stay on old.reddit.com, even if I'm logged out.

[1] https://chrome.google.com/webstore/detail/old-reddit-redirec...

Sounds like a good way to pile up your test suite with a bunch useless tests.

I'm very tired of the property investors that snatch up duplexes and turn them into quadplexes and then post them all on AirBnB with deceiving photos of the room layouts and amenities.

The number of airbnb's that I've stayed at that have awkward rooms or incomplete amenities has been increasing year over year (such as a counter that prevents a refrigerator door from fully opening due to a hasty remodel, or a kitchen with a full set of pans but no silverware). The toiletries are almost always lacking, the bed sheets are always of the lowest quality - whatever the minimum the host can do to check the box on their AirBnB profile. The photos, reviews, and amenity lists on the host's profile never tell the fully story.

This is more the norm than the exception in my recent experience.

I still prefer sublime text for opening very large files or jotting quick notes.

VSCode load times are heavily correlated to what files I had open last and what extensions I have installed that trigger because of my previously opened files.

When I freshly install VSCode on a machine, it feels blazingly fast. A couple months later when I've loaded up a few linting extensions? Not so much. Of course, the power of these extensions is why I'm using VSCode in the first place. I've yet to see an IDE that can integrate these types of features and not slow-down in some aspect, and VSCode does seem to try to stay as lightweight as possible.

VScode also often has notification prompts when I open it, either from extensions, or as popups generated from the editor itself. This doesn't add load time perse, but it does add to my mental perception of it's load time.

Wow. Talk about missing the forest for the trees.

If an interviewee has a pet operating system they've developed they can probably figure out your CRUD app.

I've had a phenomenon occur to me this week that I've experienced a few times before.

I was stuck on a coding problem for the entire afternoon one day. For whatever reason things weren't clicking and I couldn't solve the problem I had. I went to bed and as I was waking up the next morning, the complete solution came to me.

I wasn't dreaming, but I was in that hazy half-awake half-asleep mode where I'm not consciously thinking about anything.

I've had this happen a few times before where I go to bed stuck on a problem and miraculously wake up with the solution. It's a very strange (but happy) feeling.

This is literally the Auth0 first contact sales pitch designed precisely to scare you out of even trying to assess the risk of building it yourself.

The reality is, though, it's never been easier than it is today to build secure auth - yes, it's still hard to do it right - but less hard than it ever has been before. So many security features are now a checkbox on a cloud provider's web interface that would have been a manual implementation of the underlying protocol a decade ago. So many languages have matured battle-tested OS libraries for handling the especially sensitive parts like crypto.

This notion that it is a "no-brainer" decision to happily trade your customer database and a % of revenue to remove the burden of stewardship seems crazy to me.

Serializing the user's session state to LocalStorage comes to mind as a reasonable solution. Comes with it's own caveats, of course (you now have a state versioning problem), but addresses all the points you bring up pretty adequately:

1. As long as their on the same device/browser, they lose zero data. This is a pretty reliable happy-path use case.

2. The only onslaught you'll deal with is simple pageviews since everything in the trial is stored client side, which also makes #3 no problem.

Man that's funny. I saw those t-shirts EVERYWHERE for a couple of years. I think it got copycatted around quite a bit. Was yours the OG?

Yeah, it gets even worse:

https://twitter.com/SarahJamieLewis/status/13848713855379087...

CloudFormation does quite a bit of work to manage and calculate state differences. It's not CDK itself managing it (to my knowledge).

This is noteworthy because many times CDK's constraints match CloudFormation's constraints. I've noticed a few issues in the CDK github repo that a feature is incomplete because they are waiting for a change or feature to land in CloudFormation first.

Same here. What used to be a private IRC channel evolved into a Skype group chat and finally several years ago evolved into a Discord server. We game together often and even have poker nights through it. It's been a huge boon to my social life during the pandemic.

I'm really glad MS is not buying Discord after seeing what they did to Skype.

The appeal of using a static site generator is that its output is, well, static. Deployment is something simple like copying some files to an s3 bucket. If I need to host an app for the CMS portion, I'm not any better off than using wordpress or another full suite CMS. It's also not clear to me from the docs what responsibilities the content creator have in the workflow (in my experience, teaching github or markdown makes things a non-starter most of the time)

Zero custom scripts other than some vanilla JS I wrote that loads on a few pages. Your numbers are pretty similar to mine. It's not that much different than the WordPress site that replaced it, but it still seems higher than it should be.

I'd love to use a static site generator, but no simple WYSIWYG for the people that create content on the site make that a non-starter.

They have an export feature right in the settings page for your site.

I chose Squarespace for my company's marketing site, and I'd mostly agree with some caveats:

* The developer mode is poorly supported. For example: you can't copy your site if it is in developer mode (which means you can't setup a staging environment and everything you push to your site will be live instantly). The local development server is also very buggy and inconsistent with the production behavior. It feels like something that they developed early on in their road map and haven't spent much time on since then. If I could go back and do this over again, I'd pick an existing theme and just customize it instead of building out a whole theme in developer mode.

* The site is S-L-O-W with tons of bundled assets that it injects into your theme. I have no idea what any of it is doing, but it definitely increases page load.

That said, if you find yourself wanting a solid CMS that has the full breadth of WordPress like features but don't want to deal with the hosting, SquareSpace is a solid choice.

That’s not “not following toxic sources” that’s actively pruning them from your connections.

Why participate in that in the first place when toxicity grows there like a weed?

I spent a lot of time with CDK last year using Typescript. Very impressed with it and I have zero regrets after moving over our app's infrastructure to being deployed with it after 9 months in production.

I felt like being forced to use typed IaC helped me understand AWS API's and services themselves better.

It's reductive to suggest that the top 100 represents "general culture" (whatever that ambiguous term means). There aren't solely niche musicians outside of the top 100. Theres plenty of musicians with millions of listens/views on streaming media that don't enter billboard lists that have extremely prodigious careers that have complex lyrics, use a wide assortment of instruments/equipments, evolve genres, etc.

Your data might be objective, but it's still a narrow slice of a much broader ecosystem.

The top 100 hasn't been a relevant indicator of modern music since Napster. The top 100 is a record industry owned entity with a tremendous amount of self-interest.

If you can't find more diverse modern music than the top 100 offers today you're not trying that hard.

> When I need docker, I like having a native linux experience.

This is a severely underrated point. If you're not running docker on a linux host, you are running it on a VM and getting all the performance tradeoffs you would get with developing on a VM. There seems to be a lot of developers that are unaware of this.

Ever since I discovered n-gate.com HN comments read a little different than they used to...

Thanks for this, I'll keep this in mind.

Is the output suitable enough for a decent UX testing prototype?

Someone on HN once said (sorry I forgot who):

The only intrinsic value of Bitcoin is that it is an artifact of ritual sacrifice.

You're not really out of the weeds when it comes to dedicated servers either. You still need an on call resource to manage the software in that scenario.

One FTE that does on-call system engineering is wayyyy more expensive than a fleet of 20 vm's. Not to mention the cost of hardware itself.

I know there's a threshold where the cost benefit works in favor of on prem, but it's nowhere near most 20 node k8s clusters.