hipadev23

swoole makes websockets extremely easy and fast.

webRTC is a beast with the STUN server needs better handled elsewhere.

With standard opcaching and jit enabled, the reparsing step doesn’t happen.

Initializes for each request is standard with php-fpm, but swoole/workerman/etc are rapidly gaining popularity and run as a more typical app server with coroutines.

That's good to hear. I'm stuck with the email based support and apparently the low-quality crew only (my spend is about $500/mo).

Depends entirely on how much per-customer/seat you can charge. Retool is fairly expensive, but they've just recently started offering external user pricing. It's $7-10/user/mo and caps at $4250/mo at 500 users, at which point additional users are free. At 4,250 users you'd be paying $1/user/mo. They also have embedded/portal type solutions now, so your users won't know they're logging into Retool. I'm not sure if you could support a free-user tier without likely violating their TOS, though it does depend on what your SaaS offers. You can host public-facing no-login-required apps for free.

Downsides? The app overhead can be bulky. It's not something users are going to quickly open, look at some stats, and close imo. If your use-case is users logging-into the platform and likely keeping the dashboard/whatever open for hours or indefinitely? That's a better fit.

Their support team needs a lot of work. They generally are slow to respond and don't understand their own products or pricing. A lot of what should be simple questions end up taking multiple back and forth emails where you find yourself explaining the nature of your problem/question to the support person. It's extremely frustrating to the point I've thought about abandoning them over how incompetent support is. That said, the CEO is really responsive to direct emails...

Yep exactly. I’d argue they create echo chambers far more than we’d like to admit. It’s also not simply voting but also algorithmic of sorting content by likes, and implicitly excluding anything under a certain threshold.

I’ve felt this way for nearly twenty years now, coinciding with the rise of reddit, twitter, and facebook alongside the decline of forums, blogs, and rss.

On any topic with even slight amount of contention, you very quickly see people aligning toward the one-true-belief because they observe and react to the crowd supporting that opinion.

Genuine question: Do you feel like you're learning the language/frameworks/techniques well? Or do you feel like you're just getting more adept at leveraging the LLM?

Do you think you could you maintain and/or debug someone else's application?

I don’t think you realistically can. I’d instead approach it from limiting the reach of new accounts until proven as good actors.

Or switch it back to invite only, as there’s a massive userbase now, and if you invite a problematic account it becomes a problem for your account too. Operate on a vouch system.

Due to how easy it is to setup accounts and post on Bluesky, it’s likely many of the same operatives behind the propaganda and bot armies on Twitter are now pushing the same vitriolic content, and triggering these reports. If they can negatively impact Bluesky at a critical moment, it’ll reduce the flow of users who will quickly surmise “oh this is just like twitter”

He absolutely has not been “in” for 20 years. Someone will publish his actual record between active duty, active reserve, and inactive reserve. He was in the National Guard as a basic infantry officer. No ranger tab, no SF, hell not even airborne, so let’s stop acting like he’s a badass. He’s perfectly qualified for leading a company-sized unit. That’s it.

He has, in aggregate, around 4-5 years of experience as a junior officer. Most of his time is IRR which has ~zero [1] obligations and is basically just a higher priority draft list for former military.

I know you know all of this, of course, and you’re either defending him in bad faith or pure ignorance, so I’m mostly replying for the benefit of others.

[1] https://ec.militarytimes.com/guard-reserve-handbook/joining-...

> You can't even bring yourself to mention that the Defense nominee has actually served in the Army, and is a decorated veteran

The amount of copium with his pick is incredible. Hegseth is woefully underqualified to be the SecDef. He was in Army Natl Guard, not Army. He spent a year at Guantanamo in 2004, deployed to Iraq in 2005-2006 and Afghanistan in 2011-2012. That's it. The rest of the time was ARR and IRR. He has never held any public office at any level.

I'm not trying to dismiss his military service itself, it's fine, but to imply it remotely qualifies him to be the SecDef is beyond reason. He's a junior officer fit to lead a company (100-200 soliders) at best.

You don’t think a fox news host is a great SecDef? Or a non-practicing briefly disbarred AG?

Is Alex Jones the new Press Secretary yet?

A web-based multiplayer-game operating over websockets. They turned off my server a few hours into launch day.

I manage production with docker compose just fine. This article is not great.

Copilot is ridiculously easy for a CTO/IT to enable for an entire org. That doesn't mean it's providing the level of value that Cursor is. It's just like the number of "users" Microsoft Teams has, is implicitly tied to the number of Office/Outlook customers. Whether or not people actually use the tool is frankly unrelated to how many customers they can report as "using" a tool.

Incumbents (Google, Microsoft, Adobe) seem to broadly struggle at reasonable integration of AI. Just look at windows/github copilot compared claude desktop or cursor.

This is such a cute take.

> when someone tries to make metaverses happen again

While web3/metaverse as a branding concept flopped, the incumbents (roblox, fortnite, minecraft, gta creator mode) are growing strong with roblox leading the pack

Hetzner will arbitrarily null route your traffic, proceed with caution running anything important there.

Are you using s3 local cache? Do you have heavy writes? What type of s3 disk type, if any, are you using? (s3, s3_plain, s3_plain_rewritable)? Or are you just using the s3 functions.

Clickhouse is amazing but I still struggle getting it working efficiently on s3, especially writes.

Sub-millisecond? Literally nobody in large data analysis is doing that. Can you cite some sources or show me what sort of setups (with any DB tech) that you're able to run meaningful queries on S3 sources with even sub-second latency?

Think it was a just a gamble. They have lower latency edge CDNs than literally everyone else in tech, so they’re in the best position to deliver cloud gaming. And this approach enabled them to roll it out to all user devices without needing any software upgrades.

It’s a neat idea, but the games they made simply aren’t fun.

I don’t totally understand the fascination with storing analytical data on S3. It’s not fast, and if you’re in a write heavy environment it’s definitely not cheap either.

What’s with the avoidance of clickhouse or duckdb paired with insanely fast EBS or even physically attached storage? You can still backup to s3, but using s3 for live analytics queries is missing out on so much of the speed.

what was the actual cost? cpu?

> why did they do practically nothing to prevent things like this from becoming a reality

https://www.npr.org/2022/12/22/1139951463/electoral-count-ac...

The real winner is going to be a framework that, during dev, transparently materializes CTEs to temporary tables so you can iterate on them like you’re saying, while continuing to harness SQL for the end product.

Yeah. It’s difficult for true hackers to get a job because they tend to be wildly more competent than their peers and their superiors, they implicitly have a low tolerance for performative bullshit meetings and jira tickets, and they often lack the expected educational and professional credentials that hiring folks look at.

If someone wants to hire them, offer actual worthwhile bug bounties ($100k to $1M) on hard problems. And then try and hire those people after you pay the bounty.

This is cool. I hate that frotnend development is such a shitshow these days this sort of tooling is necessary. I miss the days of view-source and single style.css.

Complacency is how this shit happens.

What about WebTransport? I thought that was the http/3 upgrade to WebSockets that supported unreliable and out-of-order messaging