This leaves out the fact that attempting to make a human out of these just results in an abomination (and the ability to do alchemy without a transmutation circle).
This may be tangential, but the interesting thing to me about the Warcraft Wiki is that it serves the lore and API information in great detail and is my go-to resource for those. But when it comes to precise data about the content (e.g. spell data and its coefficients), guides for current content, etc. Wowhead has much more relevant content in greater detail - which is a shame because to me the navigability and discoverability on Wowhead is nowhere near as good as MediaWiki.
My dream is somebody takes the data from Wowhead and ports it into MediaWiki and the community rallies behind keeping that in date, but I know it's a bit of a pipe dream.
One of the best cases of leaving Fandom that I know of is the Runescape wiki mentioned in the article[0]. The community that ran the fandom wiki had buy-in from the creators of Runescape to assist with the transition, help with funding, and eventually direct integration into their games. In a game as information-dense as Runescape, that updates weekly, the wiki is basically a necessity for folks to play efficiently, or to find out how a new update actually works.
Fandom isn't the only bad wiki site though. Fextralife had (has? I haven't kept up) an issue where they were embedding Twitch streams on each page load, which was boosting viewer counts for whatever streamer they decided to embed on every page.
I'd love to see a world where more companies self-host the wiki for their game/TV show/etc, especially given the relatively low cost of deploying and hosting, but I also understand that most companies don't have the motivation to do that as it doesn't always directly impact their bottom line and it can take effort to moderate.
Unless they've moved backwards, that hasn't been doable for a very long time. The client in the past didn't have the information sent to it about things in fog of war. The closest was seeing a skill shot come across your screen and cheats gave an approximate location of where it was launched in fog of war.
I haven't kept up on the current state of cheats, but 10 years ago there were cheats that would automatically last hit minions (necessary to gain gold), automatically path towards your cursor while aiming and targeting your skills, move you out of the way of incoming abilities, all with inhuman reaction times.
There was a motto through the LoL cheating community of "Spacebar to Win", where spacebar was the keybind for unloading all your abilities and trying to kill the opponent. A player with good game knowledge and game sense would be easily able to climb to Challenger (the highest rank bracket).
There was a bot that played as a support champion and with a basic decision tree it reached Silver 2. Someone wrote a neural net algorithm for it instead and it reached Gold 4.
Is there any framework/system that distributes the work across multiple GPUs on different computers over a network (LAN or WAN)? I'm not concerned much about latency or generation time, but would love to train or load up huge models and send jobs to run overnight.
I've used coqui.ai's TTS models[0] and library[1] to great success. I was able to get cloned voice to be rendered in about 80% of the audio clip length, and I believe you can also stream the response. Do note the model license for XTTS, it is one they wrote themselves that has some restrictions.
In my experience, modern entertainment takes too much cognitive load to get up and running. With streaming/on demand, I have to make pointed decisions on what show I'm watching. I don't always want to make a decision right now of what to watch, I just want to watch something that's good enough or fits a certain theme. TV Channel creation programs, like DizqueTV[0] or ErsatzTV[1] have taken the load off, or at least lets me do that cognitive work in advance when making the channels.
You still need some way to get input to swap between these digital channels (I use Plex to surface my shows and schedule), but if you have existing local content it really feels like the "old" way of doing TV. You can even add commercials between episodes if you wanted!
I'm not sure if I'd call this foolproof for a 94-year-old, especially since sometimes the software needs to restart, but it's a step above modern streaming IMO.
I'm a big proponent of having sanity checks that values are populated properly, because all it takes is some misalignment between the backend and frontend on the API definition to break some parameters (name mismatch, added in one not the other, etc.). When I work with latitude and longitude, I typically add a test location/customer/etc. at Null Island so that I (and QA) have a very easy time debugging this particular issue. It's come in handy a few separate times!
Ruffle[0] can be embedded in your website to make flash work in modern browsers. Neopets actually did just this a few days ago[1] to bring back their catalog of old flash games.
So if you can find a way to write Flash (the old tools should still be fine, but I haven't looked too deep) you can leverage it and let folks play today.
> The standard penalty for a violation is $5,000, which is a fraction of the value of the bonus space developers receive from the agreements. For example, the owners of 325 Fifth Avenue have been assessed a total of $54,000 in penalties since 2015. By contrast, the bonus floor area that the developers gained could be worth approximately $80 million if used for residential space, based on 2022 sales prices provided by Jonathan J. Miller, a New York City real estate appraiser.
When the penalty is this much lower than the benefit, it's just the cost of doing business. I wonder if this same problem would happen if the fine was scaled to better reflect the increase in value.
I recently installed Clone Hero, with my only PC Guitar Hero-like experience being Frets on Fire probably 15 years ago, and I was blown away by how it felt exactly like Guitar Hero.
FoF felt so clunky whereas Clone Hero is a delight. It handled high FPS well, the song interface looked exactly like Guitar Hero, and the customizability is the icing on top. The community has ported over basically every song from every Rock Band or Guitar Hero game, so there's more than enough content. Clone Hero was so smooth I bought a wireless receiver to use my old Rock Band drums and play with my wife. It's a blast!
I don't know that it's a great plan to do a blogpost and in-app notification as the first round of reminders and email on the day of the change. Both the blogpost and in-app notification requires you to explicitly go on GitLab and see there's a problem. Maybe there's a reason to avoid it, but emailing from the get-go seems like it is the right move for transparency and not rug-pulling.
Super exciting read! I was interested in the Xbox 360 modding scene and re-reading this brought back a ton of memories from reading about all this stuff the first time!
Thanks for compiling this, the quality is excellent and the content is approachable.
>SQS has a many-to-one relationship. You can send messages to a queue from many different producers but only one consumer can be defined. A consumer is another application, most often some compute instances such as Lambda, EC2, or Fargate.
My understanding is that you can have multiple consumers of an SQS through the use of visibility timeouts[0]. Once a message is consumed it is as if that message doesn't exist for all other consumers until it reaches a timeout period or is marked done by that consumer. You can also manually mark a message as being ready for other consumers. This moves the message back into the queue for the other consumers to see.
I'm going to be linking this article to my team. We've been talking about moving to SNS/SQS/etc. and this article helps understand the use cases and distinctions better.
One interesting use case to me would be cataloguing features of an item you own. When you are dealing with insurance, they will reimburse you for a product of like kind and quality. For example, if you have an older consumer stand mixer that uses metal gearing instead of plastic gears (like lots of modern consumer stand mixers), then they might be forced to give you a check to buy a commercial stand mixer that does have metal gears. Adding more features typically means you will get a better payout to replace it - you get Venn diagrams of functionality that only an expensive replacement will fully cover.
I'd love to put in a product code and import functionalities of that product. The data integrity and seeding might be tricky to pull off, but it'd become extremely valuable for anyone going through insurance.
What's humorous to me about this is that both ideologies can probably make this claim. I know a lot of people who are pro-vaccine and mask mandates that are mad that this pandemic is still going on because of people believing misinformation and refusing to do their part. I know people who are anti-vaccine and anti-mask mandate who are mad that they are being "forced" to vaccinate (in effect, because of businesses adding requirements - the government itself is not forcing vaccination for the general public). Both would say the other side is causing their strife. Both are right, if you accept their axioms/premises (which I don't think is reasonable for both sides).