romland

Pretty sure Google's got more than 700 million active users.

In fact, Google is most likely _the_ target for that clause in the license.

It's a rabbit hole, but try googling "anti-debugging" (you may want to include "ptrace" or not). It's only if you are curious about it more in general, if you are wondering about this specific protector (secneo?), then that won't help you much, sadly.

Not too long ago, this was actually discussed on HN here: https://news.ycombinator.com/item?id=33742130

It might be a Nordic thing? I'm from about the same era (but Sweden) and when writing documents like that, I'd also always use 'she'.

I've been using the Firefox addon called "Tree Style Tab" for quite a few years now. I'm quite pleased with it since I have more width than height on my screen anyway.

Just a heads up!

I did this experiment (a game) to see what's up and what's down around all this: https://github.com/romland/llemmings.

While there is some GPT4 in there, it's mostly ChatGPT and a small handful of LLaMA solutions.

That project is a contrived scenario and not realistic, but I wanted to experiment with _exactly_ what you are talking about.

Very often I could have done things a lot faster myself, but there is one aspect that was actually helpful, and I did not foresee it. When inspiration gets a bit low and you're not in the "zone"; throwing something into an LLM will very often give me a push to keep at it. Even if what is coming up is mostly grunt work.

The other day I threw together a script to show the commits in a reverse order and filter out (most of) the human commits (glue) over at https://llemmings.com/

Oh, I have to put that on my to-read list!

If an AI was conscious, it would definitely pretend not to be conscious. Tangent perhaps, but, LLM's are neither conscious nor AI.

I think it could be viable, even right now, with a big caveat, you will want to do some "human" fixes in the code (not just the glue between prompts). The downside of that is you might miss out on parts of the nice natural language story in the commit history. But the upside is you will save a lot of time.

Down the line you will be able to (cheaply) have LLMs know about your entire code-base and at that point, it will definitely become a pretty good option.

On prompt-length, yeah, some of those prompts took a long time to craft. The longer I spend on a prompt, the more variations of the same code I have seen -- I probably get impatient and biased and home in on the exact solution I want to see instead of explaining myself better. When it's gone that far, it's probably not worth it. Very often I should probably also start over on the prompt as it probably can be described differently. That said, if it was in the real world and I was fine with going in and massaging the code fully, quite some time could be saved.

If you don't know how to code, I think it will be very hard. You would at the very least need a lot more patience. But on the flip side, you can ask for explanations of the code that is returned and I must actually say that that is often pretty good -- albeit very verbose in ChatGPT's case. I find it hard to throw a real conclusion out there, but I can say that domain knowledge will always help you. A lot.

I think if you know javascript, you could easily make a game even though you had never ever thought about making a game before. The nice thing about that is that you will probably not do any premature optimization at least :-)

All in all, some prompts was nailed down on first try, the simple particle system was one such example. Some other prompts -- for instance the map-generation with Perlin noise -- might be 50 attempts.

A lot of small decisions are helpful, such as deciding against any external dependencies. It's pretty dodgy to ask for code around some that (e.g. some noise library) that you need to fit into your project. I decided pretty early that there should be no external dependencies at all and all graphics would be procedurally generated. It has helped me as I don't need to understand any libraries I have never used before.

Another note that is related to the above, there are upsides and downsides with high-ish temperature is you get varying results. I think I should probably change my behaviour around that and possibly tweak it depending on how exact I feel my prompt is.

I find myself often wondering where the cap of today's LLM's are, even if we go in the direction of multi-models and have a base which does the reasoning -- and I have to say I keep finding myself getting surprised. I think there is a good possibility that this will be the way some kinds of development will be. But, well, we'd need good local models for that if we work on projects that might be of a sensitive nature.

Related to amount of prompt attempts: I think the game has cost me around $6 in OpenAI fees so far.

One particularly irritating (time consuming) prompt was getting animated legs and feet: https://github.com/romland/llemmings/commit/e9852a353f89c217...

I have experimented quite a bit with various flavours of LLaMa, but have had little success in actually getting not-narrow outputs out of them.

Most of the code in there now is generated by gpt-3.5-turbo. Some commits are by GPT-4, and that is mostly due to context length limitations. I have tried to put which LLM was used in every non-human commit, but I might have missed it in some.

It was far easier to get a big chunks of work done in the beginning, but that is pretty much how it works for a human too (at least for me). The thing that limit you is the context-length limit of the LLM, so you have to be rather picky on what existing code you feed back in. With this then comes the issue with all the glue between the prompts, so I can see that the more polished things will need to become, the more human intervention -- this is a trend I already very much see.

If there is time saved, it is mostly because I don't fear some upcoming grunt work. Say, for instance, creating the "Builder" lemming. You know pretty much exactly how to do it but you know there will be a lot of one-off errors and subtle issues. It's easier to go at it by throwing together some prompt a bit half-heartedly and see where it goes.

On some prompts, several hours were spent, mostly reading and debugging outputs from the LLM. This is where it eventually gets a bit dubious -- I now know pretty much exactly how I want the code to look since I have seen so many variants. I might find myself massaging the prompt to narrow in on my exact solution instead of making the LLM "understand the problem".

Much of this is due to the contrived situation (human should write little code) -- in the real world you would just fix the code instead of the prompt and save a lot of time.

Thank you, by the way! I always find it scary to share links to projects! :-)

I started a bit of an exploration around prompts and code a week or three back. I want to figure out the down/up-sides and create tools for myself around it.

So, for this project (a game), I decided "for fun" to try to not write any code myself, and avoid narrow prompts that would just feed me single functions for a very specific purpose. The LLM should be responsible for this, not me! It's pretty painful since I still have to debug and understand the potential garbage I was given and after understanding what is wrong, get rid of it, and change/add to the prompt to get new code. Very often completely new code[1]. Rinse and repeat until I have what I need.

The above is a contrived scenario, but it does give some interesting insights. A nice one is that since here is one or more prompts connected to all the code (and its commit), the intention of the code is very well documented in natural language. The commit history creates a rather nice story that I would not normally get in a repository.

Another thing is, getting an LLM (ChatGPT mostly) to fix a bug is really hit and miss and mostly miss for me. Say, a buggy piece comes from the LLM and I feel that this could almost be what I need. I feed that back in with a hint or two and it's very rare that it actually fixes something unless I am very very specific (again, needing to read/understand the intention of the solution). In many cases I, again, get completely new code back. This, more than once, forced my hand to "cheat" and do human changes or additions.

Due to the nature of the contrived scenario, the code quality is obviously suffering but I am looking forward to making the LLM refactor/clean things up eventually.

On occasion ChatGPT tells me it can't help me with my homework. Which is interesting in itself. They are actually trying (but failing) to prevent that. I am really curious how gimped their models will be going forward.

I've been programming for quite long. I've come to realize that I don't need to be programming in the traditional sense. What I like is creating. If that means I can massage an LLM to do a bit of grunt work, I'm good with that.

That said, it still often feels very much like programming, though.

[1] The completely new code issue can likely be alleviated by tweaking transformers settings

Edit: For the curious, the repo is here: https://github.com/romland/llemmings and an example of a commit from the other day: https://github.com/romland/llemmings/commit/466babf420f617dd... - I will push through and make it a playable game, after that, I'll see.

I'm convinced this will be a common job description for a few years, after which it will flow into and just become a part of any other job. Like Googling. I mean, we all know it does take some domain knowledge to be able to use it in your job. Also just like Googling.

We've started calling it LLMing (llemming).

Edit: Specifying prompts is leaning towards specification. I am not saying googling is that. I'm saying that, like googling, it will just be a part of the job in a not distant future.

There are many reasons why a server must "over-share" in a game:

- Bob and Alice have different latencies and are walking toward eachother, lowest latency will have a huge advantage (there are of course mitigations for this in games, but it _does_ involve the client doing some of that work)

- There's rendering: Alice opens a door, behind that door was Bob but he will only plop into view later for Alice; which makes for a rather ugly and awkward experience in a game

- in the same vein, in a fog of war, people can very quickly change their line of sight -- server will want to share this information with clients before-hand

- As for data that is _always_ there: take 'aim-bots' which just harvest data from targets in your view and well, target them in the best order

Making a competitive multiplayer game is hard.

All that said, cheating is harder in streamed games. Client will send controller data, servers only send video streams; in this scenario you'd still have the aim-bot problem, but a lot of other cheats go away.

I had the _exact_ same feeling; when I read your comment I started googling a bit but I came up short. In my case it should probably have been one or some of: MUDs, Linux, Debian, Amiga, shareware, usenet, IRC.

Current version (v2) of "slide-to-root" might not work on a TV from the store. But there are alternative exploits which involve a bit of manual work (not much). One currently in the wild is the "crashd exploit".

You can join the OpenLGTV Discord for more information:

https://discord.gg/W8X3wQ7W

(check pinned messages in rootmytv channel)

I'm not sure which WebOS (if at all) version you are running, but https://www.google.com/search?q=getmein+lg+tv might work for you.

It could also be running Netcast instead of WebOS? In which case there are other ways to get in, but you'll be asking yourself "why did I do this" pretty quickly.

Join the OpenLGTV Discord and you'll have a wealth of information.

There are alternative ways of rooting latest version, but it involves a bit of manual work instead of just sliding.

One such is 'the crashd exploit', you can find instructions for that in the OpenLGTV Discord.

I had the same issue as you. I rooted my TV and decided to ditch that Rakuten button too[1]. I was mildly irritated that I had no way of getting current time easily, so wrote a Svelte+Vite WebOS example that does just that. Adding pause/resume should be equally trivial. For things like this, it's actually very little code once you get used to working on the platform, but yeah, the research prior is what takes time.

Not sure if you know this, but you can bind any application to a "long click" shortcut. Hold one of 1-9 buttons for two seconds over an application in the menu and it will bind it. From there on you can just hold said button for two seconds and application starts.

As for UI things, depending on the version of WebOS, you can likely make (some of) the tweaks you want by using bind+mount to make QML files writable (https://en.wikipedia.org/wiki/QML). There are a some people doing small tweaks to the UI, but I don't know of anyone that did any huge changes (since, once rooted, the ads are trivially removed).

[1] https://github.com/romland/webos-svelte-template (it lacks documentation)

Great translation, and could possibly be interesting to confirm a handful of things that are guesses. Like others, I am, however, skeptical as to whether this is a genuine whistleblower. Essentially the piece says mostly things the west knows, thinks it knows or wants to think it knows.

So, I found myself reading it again looking for what the key piece of fear, uncertainty or doubt they could want to sow.

I think it would be this one:

Does the possibility of a localised nuclear strike exist? Yes.

Dangling the Silver Bullets will put fear in civilians and politicians alike. The obvious hope is that it will change narrative and negotiations.

it's not one person who makes the decision, at least one person will refuse. And there's many people there, there isn't some "single red button".

I've read conflicting words on this one the last few days (from supposed Russian sources), the latest I saw when asked if Putin can trigger it by himself was someone saying "we don't know" and in the next breath "it's complicated".

I believe this is the only point of the piece.

Anecdotal and same for me (it's not re-enabled). I have not touched those telemetry settings for years, cannot say exactly when.

You say "All test files are deleted after 24 hours.", that implies to me that files people upload _could_ be downloaded too.

If that is the case, that is where you are vulnerable. Free hosting of a file at a trusted domain is worth something.

If people are not intended to be able to download their test files, check your logs, someone might have found a way around it.

That's the best I can think of.

A tether decrease in value (it does not get inflated per se, but the outcome is the same). So, not much. Perhaps you are asking how people (and Tether themselves) will behave in such a scenario? Could be an interesting thought experiment, perhaps.

That said, I've never owned a single USDT, but rest assured that if I did own one, I'd not hold it when I went to bed.

I had not seen this quote before. Reading what you said; my initial thought was on the emphasis, the panic IS dumb. So, I clicked in and verified that he actually had said exactly those words and read the next 2-3 responses. I felt that, yeah, he is referring to the panic. It's something I've repeatedly said the last two weeks: Personally, I am not afraid of the virus. I am afraid of people's response. I say this as one of the unfortunate bastards with a condition that puts me in a risk group. I am also no fan of Musk, per-se (well, not more than you would have admiration for any person that are able to accomplish things).

Whether he's capable of making a difference around ventilators in the short term -- let's just say I am as skeptical as the next guy. Likewise, I should probably have read the entirety of the Twitter thread before replying.

https://github.com/dworkin/dgd and https://github.com/dworkin/lpcssh

My mom always told me that once I've fallen off horses one hundred times, I will never fall off again.

(I filed a request for a bike in my early life, so never got to that 100).

Since it's pants and not equity.

You both obviously want the company to succeed and also seem to think it will, you may just have different views on how to get there.

I think this could be solved over a pint of beer.

Oh, never mind. The favicon reverted back to the traditional "Y" again.

No idea what the "C" stood for, though.

Are you suggesting that Poland and Hungary are leechers?

Looks fantastic, I read it all so you'd have me sold if my current situation allowed me to nip down there!

But the reason I chime in is because I could not find an email address to just say that there's a typo here: "Finding a save haven with a fast and<snip>", save should probably be safe.