BiteCode_dev

Yeah let's block port 80, too many people expose unprotected api.

Also the results given it the article are still worse than redis with zero work and no risk of breaking later, both in nb of req and latency.

And you can move redis to another server next later if needed to split the db and notification load.

I don't know, I tried kagi but it doesn't seem to have superior results to even ddg. It's even worse for non english content.

ChatGPT is better than their agent at everything, firefox extensions can filter domains and ads out of the search results while preventing tracking nand I don't even have to create an account for any of those.

Not to mention I actually mostly use search engines to for stuff that are not fuzzy anymore (since discovery with them are now terrible anyway), so I don't get that much value.

I really want to like kagi, it ticks all the boxes, but it misses a killer feature that would make me use it instead of alternatives that just do what I need.

I downvoted up out of reflex, then read till the and went back to upvote.

Courageous game you are playing here.

We spent 20 years talking about agile precisely because of this, and people took it, turned it into a checklist based processed, forgot about the short feedback loop it's precisely supposed to bring to the table to figure our your unknown unknown quickly, and declared it useless.

And now we are back to square one, where we argue again "but just make good specs".

The world is a silly place.

No, it's incentive. Always has been. Always will be.

Awareness may (or may not) help with incentive, but by itself it does jack all, as any nerds that have been actively showing their relatives how they are tracked will tell you.

Snowden's revelations did diddlely jack for the common person, nor wikileak's. Knowing smoking is bad never prevented smokers to smoke, knowing exercice is good is not enough to make somebody do it.

You need more: social pressure, conformism, dopamine, lower friction, etc.

Safety belt didn't become common because the average person were aware of the benefits. People screamed about it, complained, refused to use them.

It's the push from insurances to get them into law that turned into car manufacturer making it a feature, cops giving away fines, and social contracts pressuring you into using them that made them ubiquitous.

The myth that awareness is what is missing is one of the reasons so many topic never move because it's talking on the internet about it is easier than doing something about it. So people choose awareness as the virtue signaling easy way out of their guilt.

You may do. And I may do.

But most people won't.

See how they react to terrorism, despite the fact they amount to almost nothing compared to any other cause of death.

People are NOT rational.

All the answers are smart sounding and completely avoid to address your main point. Same with most doctors I asked it to.

They got 1 billion monthly active users in 2025. This is a write intensive workload with signaling accross all of them, including in rooms with hundreds of users, who can upload huge videos in an unlimited searchable history.

Then you got misbehaved scripts, ddos attempts, authentication for every message, encryption...

Yeah, it's expensive as hell, even with an optimised erlang backend.

In fact, if they manage to get all that down to one cent per user per month, which is very cheap for such a service, it's still 10 millions to run just in hosting.

Common python strings are interned to speed things up as well. Maybe this way of interning could benefit cpython.

I always wondered how telegram could afford being free. A chat is expensive to run, a chat with gigantic media quotas even more, and they don't have ads.

There is no way their premium plan cover their cost, espacially with their extensive bot API that multiplies their traffic tenfold.

And now they add free hosting of bots on top?

How the hell are they doing that?

With wechat in china, pix in south america and wero in europe, I think it's quite the opposite. All continents want to see visa and mastercard dead so they can bring back those transaction fees in house and it will work as the phone is the low friction payment method now.

Wero in the EU is fantastic, albeit still very young. Once it's mature and deployed everywhere, I see no reason to use something else in my country.

In fact, in France, the CB system endured despite visa and mastercard for 4 decades, so we know how to do it already.

For me, claudism is Americanism. Overly friendly despite having a very superficial relationship, careful to be as neutral as possible when giving negative feedback or opinion, superlative on mildly interesting things to make it sound exciting to the point nothing has a scale anymore, and use of jargon you heard your smart friend say to appear you know your stuff (purple DB syndrome).

Overall, I'm used to it, having worked with the US for so long. But as soon as I hear or read it, I get into transactional mode; my human side is completely gone. I assume nothing is personal, and ignore my (usually quite high) empathy signals.

This was sad with other people and I assume one of the reasons extremes are taking off (at least it feels real), but it's sane with robots, so for claude it's fine. And it's also perfect for people interacting with you with claude as an intermediary, since they clearly see it as something that needs to get over with.

Two things can be true.

And clearly, you think billion-dollar companies making dark patterns, ignoring web standards and choosing to track people left and right are less to blame the inconvenience of a banner (that warns you they do) than the people trying to protect your privacy and did it imperfectly.

I have made enought web sites and app that don't have a banner to know it's perfectly possible, even today.

I have implemented DNT support and know it was a great solution before it was taken away.

I have worked with enough clients to know why they chose the banner anyway.

Unlike you I actually read the law, and worked at implementing it. Including with and without a banner.

So I have to conclude you are not an honest actor in this debate, and you are clearly angry as well.

So I'll leave you at that.

No need for a conspiracy. There was the DNT standard, companies made more money from not taking it into consideration.

That's it.

They chose the annoying banners, then they chose dark patterns on those banners.

It is a choice.

At best they were lazy and greedy. At worse they were malicious.

It's another one, I actually didn't know about this one. Saving.

There is 15 years of documentation of this. This is explained in countless threads in HN.

I'm tired boss.

Because if enough idiots do it, stupid managers think it's the standard. The EU Commission is unlikely to know anything about their website, it's made by some department that blindly follows a manager's order that just looked at what others did and copied it.

We could all have used the DNT header as a bypass when the GPDR came out, and you can still use cookies for non tracking purposes without any banner.

I did, but that doesn't change AT ALL my initial point.

People still buy HP printers.

It's still a popular brand.

If illegal actions are just reverted and compensated by a fine, then they are just legal for a fee

Oh no, I remember well. At some point, everyone at the software home button on their phone setup, because the physical button was sometimes working and sometimes not.

Apple issued a statement saying it was a hardware failure and there was nothing they could do.

A hacker later on proved they were lying by patching the software and showing the problem went away.

That's why I'm scrapbooking every article about the trump administration right now. This time period is so wild people will doubt it really existed.

I will not be gaslighted again. This world is crazy, and people have a terrible memory.

The GDPR does not say that controllers must always delete personal data on request. Article 17(1)(a) — erasure is required only when:

"the personal data are no longer necessary in relation to the purposes for which they were collected or otherwise processed"

https://gdpr-info.eu/art-17-gdpr

It's like the cookie banner all over again. This law never, ever required a cookie banner.

The big companies are master in malicious compliance that benefit them, and let them blame the EU for it.

Rules of thumbs, international billion dollars company should be assumed to be the ones being the bad guys until proven otherwise. They have lost the benefit of the doubt decades ago.

And consumers will cry about it for about 5 minutes, then go back to reward the company.

When HP started making printer cartridges that expired even when they were still full, people complained—then bought more.

When Microsoft let the web stagnate with IE6, people complained, then turned around and did the same thing with Chrome.

When Apple deliberately put a bug in the iPhone that caused the Home button to fail, pushing people to buy the next model, people got upset—and then bought the next one anyway. I'm amazed nobody remembers that one; it was such a huge deal at the time. And there is not a single link to articles about it anymore.

When Adobe switched to mandatory Creative Cloud subscriptions, plenty of users protested, but most professionals stayed.

When Amazon remotely deleted books from people's Kindles (including 1984), it was a scandal for a month, and then... nothing.

When we found out PRISM existed, users were worried for a few months, then went right back to filling those platforms with their personal data.

When Google allowed fraudulent DMCA takedowns, shut down accounts with no appeal, and censored its search engine, there was a brief outcry, then it was back to business as usual.

When Sony put a bloody ROOTKIT on its music CDs (!!!!), people grumbled for five minutes and kept giving them money.

These companies have no reason to stop. We never make them regret anything.

I should make a website to save those for posterity, so that at least we have a track record of all the things they get away with because we let them.

We're screwed—and we deserve it.

And it's a damn shame because we have awesome hosting in the EU.

OVH is great value, Scaleway can handle lots of traffic, hetzner is so cheap...

So it's "nobody gets fired to buy IBM" all over again.

It's not, not everything need to be a single word, because the world is full of nuances.

Calling everything fascist, nazis, communists, etc. is making actual fascism, nazism and dictatorship more likely.

Because you can't raise the attention of people to the absolute priority those needs when the time come if you just wasted it on stuff that were not it again and again.

We are crying wolf, and we'll pay the price.

I actually opened a yandex mail account 15 years ago because, since I was going to be tracked, at least it would be from people who have no friendly contact with most gov entities and companies in my society.

Sad.

We have, we just have much, much better conditions for food, hygiene, personnal safety and medicine.

But have worse hormonal health (pmc.ncbi.nlm.nih.gov/articles/PMC7063751/), and are less fit (pmc.ncbi.nlm.nih.gov/articles/PMC4033061/). The flynn effect also seems to decline in some parts of the world: www.sciencedirect.com/science/article/abs/pii/S0160289619301679

It just doesn't compensate the immense gains tech created.

Turns out it's ok to be weaker if you don't have to worry about dying of parasites, malnutrition, cold.

Which, you could conclude, means the individual is weaker, but the species is stronger.

It used to be the polite thing to disclose that you used a translation app. In fact, traditionally, you disclose when you translate anything so people know the context in which to interpret your text.

In the same way, I wanna know if a book is written by some famous people just ghost written.

Of course, the point is moot. Somebody using AI to write a blog post is unlikely to be self conscious enought to thing it's necessary to disclose it in the first place.

For all its faults, this is one of the things the Python typing system gets right. It's dynamically introspectable at runtime, so you can define type, parsing and validation in one go with stuff like pydantic.

I had one machine on my entire client's network that could never download a LibreOffice document without it being corrupted.

Turns out it was kaspersky intercepting network calls, and deciding it was a very dangerous piece of file, and it would truncate it completely silently.

After wasting a non-billable afternoon on it, I just disabled the antivirus out of desperation and figured it out.

The solution was to generate a self signed certificate and TLS the connection and prevent the bugger to MITM us.

Since this day, even on a local network with behind a proxies and using a VPN, I still use https for all the services if I'm allowed.