See Directive 2019/770 for the definition of "functional" in this context. It's already a codified term: The digital content or service must be able to perform its intended functions reliably, meet the quality and performance expectations set forth in the contract, and be compatible with other systems as expected by the consumer.
Apple didn't manage to convince the rest of the industry that it's the better connector, so USB-C came out ahead. That was not the mandate. The mandate just told them to collaborate and standardize around one connector.
Apple charges money on a license for lighting cables. And it's not cheap. Surprise, their competitors don't want to pay out millions to Apple for some crappy charging cables.
People have this idea lightning stuck around because it was good. No, it was a money printing machine that required no lubrication from Apple. Even the dumbest of the dumb MBAs wouldn't have killed lightning. From a business perspective it's gold.
Funny thing is server shutdowns are rarely as "planned" out as you'd imagine. Look at GameInformer as we speak. In less than 1 business day the staff was kicked out with no warning, its 20+ year old website taken down, and its twitter handle deleted, hours after making a goodbye statement. This was clearly not planned more than a month in advance by Gamestop.
Planned Obscolecense implies (if not, is outright defined) as designing a product to fail in a specific amount of time. No game service worth its salt is designed to shut down in 6 months (you can probably find some chinese asset mills to prove me wrong. so take a charitable interpretation of "worth its salt")
There may be premature shutdown but no buiness plans to run servers for free forever. Ongoing server costs will be part of the business plan from the start.
I feel like it's not broad at all, it just demands that the company delivers what they advertised, since they don't usually openly disclose the temporary nature of your "purchase", because it would hurt sales. (No, hiding it in the EULA doesn't count as per EU courts.)
- Disagree on it being technically infeasible. It's basically trivial: You're probably already running the Kubernetes config on the cluster anyway. Just release the server binaries/config/docs. Laws are also usually not retroactive, so negotiating licenses that allow for this in the future seems trivial too.
- You don't lose IP by distributing anything, just like you don't lose it for distributing the client. I don't own Ford because I bought a Ford car. The only thing licensed IP in a product does it that you can't sell it anymore after it expires. It has no effect on previously sold copies.
- EULA/ToS is invalid if it contains unfair/unexpected clauses. They like to call it a service, but that doesn't mean that it actually is, legally. As opposed to SaaS, games are sold as a product with no expiration date. The EULA/ToS also always contain clauses like "terms can change at any time for any or no reason", which is inherently invalid. So the whole EULA/ToS could be invalid on its face too.
- This is just about basic ownership rights. If it's a rental/service (with a disclosed price for a specific time period), then it's fine. Otherwise it's a product and you have to abide by the regulation for products. Anything other than these two options is inherently unfair, because you can't assess the value of something if you don't know how long it may be used for.
>just demands that the company delivers what they advertised
I don't know any company that advertises "game will stay up for X amount of time and will always be playable". On the contrary I'm pretty sure every EULA specifies that they aren't. you're logging into someone else's server, so I should log in knowing that server won't say up forever
>EULA/ToS is invalid if it contains unfair/unexpected clauses.
how is it unexpected 30 years into the internet that "oh yeah, this is a server-based game, it won't stay up forever". It's unfair, but laws are rarely made with a goal of perfect "fairness".
> Disagree on it being technically infeasible. It's basically trivial:
Nothing is trivial in tech. Not unless you're talking on the scale of years. This isn't even a gamedev thing, it's just that there's always random footguns and pitfalls due to the nature of shifting to a attrition strategy instead of a retention strategy.
> You don't lose IP by distributing anything
if you distribute IP you do not own, you end up taking damages while also having the game taken down. very few games are made fully in house anymore.
It stops being their property when they sell it. They keep owning the IP, but not the individual copy. Anything else is just someone coming to your house and stealing something you bought back.
Respecting basic property rights is not up for debate. If a business model relies on violating them, then it probably doesn't deserve to exist in the first place.
The problem there is that games don't usually advertise a subscription, like basicall all SaaS products do. They masquerade as a good for a one-time payment, then turn around and pretend it was a rental all along.
Note that it's not about running servers for all eternity. It's about patching out the requirement of an _official_ server and/or releasing dedicated server software, at least _after support ends_, like games have done for decades already.
MMOs can be both. If it quacks like a good, it is one, no matter what they say in the ToS/EULA. Stuff like World of Warcraft would likely be unaffected, because they are up-front about the duration you pay for.
Seconded, I've tried posting videos here before. His explanation of the great pyramid being a public/private devotion place, not a secret grave, makes the most sense to me. It would be like Lenin's Mausoleum. Everyone knows where its at, who is there, and you can go see him and leave offerings.
Previous tombs were robbed and looted because the king was buried and forgotten and no one cared anymore. Probably helped their followers maintain power after their death too.
(purposefully not using any names, I am skeptical on the official story of who built what for who)
It's pretty hard to draw this line in Python because all built-in types and functions are effectively C extensions, just compiled directly into the interpreter.
Conversely, you can have pure C code just using PyObjects (this is effectively what Cython does), with the Python bytecode interpreter completely out of the picture. But the perf improvement is nowhere near what people naively expect from compiled code, usually.
Yes, which is why I would argue that IO is a particularly bad benchmark here, since everything is just a thin layer on top of the actual syscall, and those layers don't do any real work worth comparing.
The only thing that makes sense to compare when talking about pythons performance is how many instructions it needs to compute something, versus the instructions needed to compute the same thing in C. Those are probably a few orders of magnitude apart.
