No one needs to deduplicate over a longer period than a few minutes, or a single session. If you need that, then you're doing something shady. If a user visits your site, clicks a few things, leaves and comes back two hours later, you don't need know if it's the same person or not. The goal of analytics is to see how people in general use your website, not how an individual person use your website.
So just take IP address, browser details, your domain name, and a random ID you stick in a 30 minute session cookie. Hash it together. Now you have token valid for 30 minutes you can use for deduplication but no way of tying it back to particular user (after 30 minutes). And yes, if the user changes browser preferences, then they will get a new hash, but who cares?
> No one needs to deduplicate over a longer period than a few minutes, or a single session. If you need that, then you're doing something shady. If a user visits your site, clicks a few things, leaves and comes back two hours later, you don't need know if it's the same person or not.
Sure you do if for example you want to know how many unique users browse your site per day or month. Which is one of the most commonly requested and used metrics.
> So just take IP address, browser details, your domain name, and a random ID you stick in a 30 minute session cookie.
That looks a lot like a unique identifier which does require a user's consent and a cookie banner.
> Now you have token valid for 30 minutes you can use for deduplication but no way of tying it back to particular user (after 30 minutes)
The EU Court of Justice has ruled in the past that hashed personal data is still personal data.
> And yes, if the user changes browser preferences, then they will get a new hash, but who cares?
It will also happen after 30 minutes have passed which will happen all the time.
> Not rocket science.
And yet your solution is illegal according to the GDPR and does still not fulfil the basic requirement of returning the number of unique users per day or month.
So take the IP, browser agent, your domain name and some other browser identifiers, stick them together and run them through SHA3-256, now you have a hash you can use for deduplication. You can even send this hash to a 3rd party service.
Or assign the user an anonymous session cookie that lasts an hour but contains nothing but a random GUID.
Or simply pipe your log output through a service that computes stats of accessed endpoints.
I think this scheme still requires consent since you are processing pseudo anonymous identifiers that fall under personal information without the essential function basis. Hashing is considered insufficient under the GDPR iirc. Have you asked a lawyer about this?
We have plenty of tech companies. The reason you've not heard about them is because most of them cater to their domestic market first. Neighbors second. Rest of the world third or never.
Europe has a shitload of homegrown competitors. The problem is that users here in Europe either goes for a national service or for an US service. They don't look up what their EU neighbor has to offer. In fact, most don't bother translating their services to appeal to the entire EU market.
If you live in country X, you will only ever learn about services from country X or from the US. No one here knows what goes on in neighboring countries.
It's easy to think the EU is like the USA, but it's not, it is still separate sovereign countries with their own language and culture.
I think there's something like 24 national languages in the EU. I can hardly blame hetzner for not translating their services to say polish and think it's entirely the wrong approach anyway.
It's really true language is a big barrier but honestly the solution cannot be for every single company to offer services in 20 languages. It can't be. English must be adopted.
I cringe when I read this. Why not German? There are more native German speakers than any other language in the EU. Also, in the age of LLMs, translating (on a best effort basis) to (at least) 24 different languages is trivial.
Now it should be clear why one is better than the other. The shared language of most is English, so you have the least amount of "extra learning" required.
Also, the number for German is generous in that it includes people that speak wildly "incompatible" dialects and accents. While people in Bavaria technically speak "German" and having them talk to other people that speak "German" (with various dialects) is easier than asking either to speak English as their primary language, that doesn't really solve the problem of even intra-German language rivalry.
Of course one thing will unite Bavarian and Saxon and Swiss and Austrian German and other highly accented/dialectic German speakers: They'd rather speak "German" (and deal with weird pronunciation/words) than English as an official language ;)
I have asked multiple native German speakers about the "linguistic distance" between various styles within the Federal Republic of Germany. It is completely overstated that people don't understand each other or are "annoyed". All German children grow up learning standard German in schools. Yes, they may speak a different at home and in the community, but they are all fluent in standard German. I am pretty sure most standard German speakers can communicate clearly with all of Germany, most of Austria... and Switzerland is a roll of the dice. Still, anyone in the German-speaking half of Switzerland that is university educated will surely speak standard German. Again, they may speak Schweizer Deutsch with their family and friends, but can also speak standard German, especially in a business setting.
Are there more distinct markets in the EU/EEC where adopting german would give you a quantifiable economic and/or competitive advantage over adopting english?
Why work work with native language rather than spoken? According to wikipedia less than 20% of the EU is a native German speaker while 47% speak English. When talking about technical people who may be looking into something like Hetzner it is probably higher than 47%.
I never really looked at it that way, but I think you're right.
Although, non-European-owned companies aren't necessarily incentivized to look towards European companies.
Looking towards your European neighbors mostly comes down to logistical situations. In those sectors, multilingual services are more common.
I don't think XSLT was invented for the purpose of rendering XML into HTML in the first place. Perhaps it never should have been introduced in browsers to begin with?
XSLT was invented to transform one XML document to another XML document.
Browser can render XHTML which is also a valid XML.
So it's pretty natural to use XSLT to convert XML into XHTML which is rendered by browser. Of course you can do it on the server side, but client side support enables some interesting use-cases.
Look up how many refueling launches are required and you'll see the problem, especially because no matter if Elon says so, the upper stage will never be reusable, even if caught.
Every moon mission will require that they pre-build a HLS and probably 15 full stacks.
> Look up how many refueling launches are required and you'll see the problem, especially because no matter if Elon says so, the upper stage will never be reusable, even if caught.
The Space Shuttle was reusable, and SpaceX is using an improved variant of the Space Shuttle heat shield, so it seems quite certain that Starship will be reusable. The question is more: how much refurbishment will it need? The Space Shuttle required extensive amounts. SpaceX will likely be able to improve on that a lot, though it isn't clear how long it will take.
You're welcome to believe that. Visible progress to date suggests otherwise to me; I pretty much ignore what Elon says as much as possible. Besides, however ridiculous, 15 full stacks would still be cheaper than a single SLS launch in all likelihood.
Even if I'm wrong, though, it wouldn't invalidate the point I'm making in this thread: BO's Mk2 has the exact same issues in a more complex architecture.
