> we recommend keeping task-specific instructions in separate markdown files with self-descriptive names somewhere in your project.
Why should we do this when anthropic specifically recommends creating multiple CLAUDE.md files in various directories where the information is specific and pertinent? It seems to me that anthropic has designed claude to look for claude.md for guidance, and randomly named markdown files may or may not stand out to it as it searches the directory.
You can place CLAUDE.md files in several locations:
> The root of your repo, or wherever you run claude from (the most common usage). Name it CLAUDE.md and check it into git so that you can share it across sessions and with your team (recommended), or name it CLAUDE.local.md and .gitignore it
Any parent of the directory where you run claude. This is most useful for monorepos, where you might run claude from root/foo, and have CLAUDE.md files in both root/CLAUDE.md and root/foo/CLAUDE.md. Both of these will be pulled into context automatically
Any child of the directory where you run claude. This is the inverse of the above, and in this case, Claude will pull in CLAUDE.md files on demand when you work with files in child directories
Your home folder (~/.claude/CLAUDE.md), which applies it to all your claude sessions
I wish they would at least include the japanese cc subtitles. They must already exist so I don't understand why they can't just be packaged along with the other subs.
I've only used LLMs with Elixir, so I don't have any other experience for comparison, but I've found that although Claude frequently employs the wrong approaches in Elixir, I usually know when he'll have trouble and just ask him to read pertinent documentation first. So long as he's read the manual he seems to do just fine.
I used it a bit. The performance was exceptional. I enjoyed playing Cyberpunk 2077 on my phone with a Razer Kishi controller. Although I'm not surprised Stadia struggled to find traction, I think this is mostly due to Google's ongoing struggle with entertainment branding, and despite this, I do agree with their sentiment that streaming is the inevitable future of mainstream gaming.
Exactly right. This is it right here. It's openly hostile to users, especially considering the broader context of Weeny's pejorative comments about his customers.
That's all well and good, and if you like EGS, that's your prerogative. My response is to a question of why people don't like EGS. The point is not even that Steam is better (it is); instead the point is that nobody likes to be forced to use a platform they don't prefer, and this is made especially onerous when the forced platform is so very, very inferior.
Because it's a terrible client with virtually no features, yet their business tactics strongarm players into using it. Nobody likes to be forced to use a certain product. EGS doesn't even offer a fraction of what Steam can do. Here's what I can come up with off the top of my head; I'm sure others can chime in with more.
Steam has a robust marketplace, with user reviews and a slew of sorting and shopping cart features.
Steam has friends, chat, voice chat, and group chat.
Steam has built-in, turnkey stream broadcasting of games.
Steam allows friends to drop in and play a game cooperatively with you, even if they don't own the game too.
Steam has remote play with friends.
Steam has remote streaming to devices around your home, and big screen mode so you can plop on the couch and enjoy an optimized UI and a gamepad.
Steam has robust gamepad support and configuration.
Steam has industry-leading Linux support.
Steam has VR mode, with custom spaces, interfaces, and a market of useful tools and mods to enhance the experience.
Steam has a music player and a built-in web browser.
Steam lets you add non-steam games to your library for convenient access.
Steam has customizable user profiles and fun social features, like card trading.
Steam has experimental mods and features via Steam Labs.
Now, ask yourself, how does it feel when EGS coerces players into using their inferior storefront through strongarm business practices.
Did you create a throwaway account just to call me a dipshit? For asking an earnest question? I was just interested in why the comment was being voted down, because I'm interested in learning more about the community's standards and moderation policies.
I'd also like to know what my comment was flagged for, but I guess that's out of the question too. It's not very fun being censored with no explanation whatsoever.
Why did you jump to so many conclusions and resort so quickly to namecalling?
It's not actually hard at all; instead it's just that the easy methods are not well known, since most crypto news is all about crypto investments and get-rich-quick schemes. Easy to use payments systems are absolutely in place. lobstr, for instance, looks dead simple to me
In my experience the documentation around Elixir has been ideal. People are modeling the absolute best of behavior, generally. The popular auth libraries (like guardian) are a bit of an exception in my opnion, for some strange reason. (Pow does a good job, though)
The problem is previously we didn't listen to victims, and allowed them to be silenced. Now the pendulum has swung the other way, and we are listening unconditionally, but (occasionally) bad actors are taking advantage of our good intentions. How do we find balance?
A deed does not refer to a specific thing. It only does so in daily life as a matter of convention. It is the authority of the banks or the government or whatever that gives your particular deed any weight at all. A deed, like a dollar bill, is a piece of paper with no value. Indeed, a photocopied dollar bill has no value whatsoever. It is only our collective agreement to pair value with officially recognized papers that grants those papers value.
The situation I described does make sense, because it's exactly what we are circling around here. Anyone can sell anything; it is merely a question of whether the sale will be recognized by the parties that matter. An NFT is necessary because it distributes the transfer and sale of things without a central intermediary. There is and always will be a required authority to recognize the value of a given NFT, however.
Note that these two ideas of central authority are not equivalent. In one hand, you have a centralized authority over trade; in the other, an authority over value. Note also that in the case of NFTs, there is no "one central authority who governs all value", but instead many authorities, i.e. the creators! The artist or creator of a given work is free to recognize the NFTs that grant ownership of their creations. It is they who have the power of authority.
I have no misunderstanding about this whatsoever, in fact. I understand that NFTs confer no rights, which is why I posited the idea of a necessary authority who can declare which NFTs are valid representations of ownership of their works. It is the pairing of NFT and authority which grant the possibility of ownership.
Many people immediately jump on the idea that NFTs cannot confer ownership, and thus are worthless as a concept, but they fail to recognize that it is only one half of ownership, just as a deed to a house is just a piece of paper (that anyone can photocopy) without an authority to recognize it.
It totally disagree. I think it is quite equivalent. The difference is that we don't have a centralized authority to recognize valid NFTs, as we do for deeds. But if we did have a central authority, like perhaps the artist responsible for the work protected by the NFT, and if the artist made it clear and public that only the NFTs she recognizes could be considered valid proof of ownership of her works, then indeed an NFT would function like a deed.
Why should we do this when anthropic specifically recommends creating multiple CLAUDE.md files in various directories where the information is specific and pertinent? It seems to me that anthropic has designed claude to look for claude.md for guidance, and randomly named markdown files may or may not stand out to it as it searches the directory.
You can place CLAUDE.md files in several locations:
> The root of your repo, or wherever you run claude from (the most common usage). Name it CLAUDE.md and check it into git so that you can share it across sessions and with your team (recommended), or name it CLAUDE.local.md and .gitignore it Any parent of the directory where you run claude. This is most useful for monorepos, where you might run claude from root/foo, and have CLAUDE.md files in both root/CLAUDE.md and root/foo/CLAUDE.md. Both of these will be pulled into context automatically Any child of the directory where you run claude. This is the inverse of the above, and in this case, Claude will pull in CLAUDE.md files on demand when you work with files in child directories Your home folder (~/.claude/CLAUDE.md), which applies it to all your claude sessions
https://www.anthropic.com/engineering/claude-code-best-pract...