rpdillon

This is closer to the reality I'm seeing. Software engineers are well positioned to understand and deploy LLMs to solve problems, even in adjacent domains. One of the engineers I work with gave a short presentation yesterday. It was quite simple. He'd just taken a series of screenshots from our product and put it into a Claude desktop session with some pretty significant prompt explaining how the screenshots related to a particular user flow and what the desired changes we were thinking about making were.

Took about 45 minutes and produced a working mock that we could then play with and see if the UX felt right. It was surprisingly high fidelity, and while the code was entirely throw away, it helped a room full of people make decisions about how we wanted the tool to feel within the first couple of hours of thinking about the project. Designers bring a wealth of UX thinking to that discussion, but the process of creating the mock was done entirely by engineering.

OMP provides a significant amount of structure to the work. That costs some tokens in terms of the system prompt. When I use these systems, I balance the return of the extra guidance in the system prompt with the cost of the context. You've talked about how much context it takes, but you haven't talked about any difference in behavior.

omp offers some nice tools, like /shake, to manage context efficiently and offset some of that consumption.

Not sure about your point with command keys. OMP has all kinds of shortcuts and is highly configurable. It seems very unlikely that you cannot achieve what you're looking for in OMP in terms of shortcuts. Specifics would be immensely helpful here.

And yeah, you'll need to run your inference engine separately from your harness, for the same reason the web server and the web browser are different applications.

Yeah, saying it kills the minimalist bit misses the point of omp being a separate project. It took a minimalist-base of pi and added tools that make it usable out of the box. Hence batteries.

One example here is that it added MCP extensions to make the harness usable with MCP servers long before Pi figured out that that was the right move. A lot can be said about development of MCP over that timeframe, but I appreciated the bias towards adoption, since progress is driven by empirical experimentation rather than theory in this domain right now.

I've long championed assembling developer tools yourself from scratch and I've been an advocate of vanilla Emacs and customizing it yourself as an ideal developer experience (I realize this opinion is controversial). But OMP is essentially the VSCode to pi's Emacs.

I ended up choosing omp as my daily driver, mostly because I see the velocity of change in AI development being far too fast for me to make a worthwhile investment into customizing the environment myself, since that investment will have a relatively short half-life. With Linux, which I learned more than 30 years ago and still believe to be a fantastic investment, I can largely use the same knowledge I gained then to administer systems today. I fear that much investment in the specific tools that I add to my pi harness will expire before 2027. So I'm sort of drafting OMP until things settle down.

See if you can replicate this inside of tmux. It might be screen's escape code handling.

Yes, strong Hammacher Schlemmer vibes. I noticed years ago that they try to make every product sound like the pinnacle by prefacing it with "The". Somewhat effective, until you catch on.

https://hammacher.com/

Its easier to understand when you see the progression, IMO:

Haiku -> Sonnet -> Opus -> Mythos/Fable

Luna -> Terra -> Sol -> Astra

Weirdly it seems like OpenAI originally just released by version numbers for their models and have switched to Anthropic's naming approach of late.

> Other than Google, every other browser does that by default for free, and even Google offers chromium for free.

We disagree about the facts on this, and it's a waste of time to argue facts, but I do appreciate the discussion!

Local AI has completely changed my relationship with computers because now I can build utilities, tools, scripts, and get debugging help with my systems with far less friction than I ever could before. I used to do all of this by hand, but now I can spin it off as essentially an async sub-process to the AI that I can return to a few hours later and have the job done. Everything from system upgrades to custom Steam Deck launchers and ad hoc service scripts for my home lab have benefited from being able to run a two-bit quant of deepseek on my Strix Halo.

I think it's going to remake our relationship with software because it completely alters the economics of creating one-off utilities, whereas software up until this point is generally regarded reuse as one of the highest virtues.

We're not going to see that impact for a while though. It'll take years. App stores will eventually become on-demand software and we'll view the marketplace analogy (where human developers offer their wares) as antiquated. I'm not totally confident in this prediction because requirements gathering is still a challenge, but for anything vaguely common, I think it'll be a workable model.

Mostly just focusing on positives in my life, not particularly bullish on the industry generating 6T/y.

Your first paragraph is well-taken. Even if it were infringement, infringement is not theft, as HN likes so often to assert.

The Supreme Court agrees.

> Since the statutorily defined property rights of a copyright holder have a character distinct from the possessory interest of the owner of simple “goods, wares, [or] merchandise,” interference with copyright does not easily equate with theft, conversion, or fraud. The infringer of a copyright does not assume physical control over the copyright, nor wholly deprive its owner of its use. Infringement implicates a more complex set of property interests than does run-of-the-mill theft, conversion, or fraud.

https://supreme.justia.com/cases/federal/us/473/207/

It's about alignment of interests. They believe that there is a business that is valuable selling a user-centric browser. A user-centric browser does not exist today.

Google Chrome is a frog-boiling exercise where they slowly take away freedoms to run, for example, ad-blocking in the browser and they do so under the name of security. They've already tried to push the web integrity initiative. They'll try again.

Mozilla, as I've already pointed out, is criminally mismanaged in the sense of poor allocation of resources towards building a better browser. On top of this, they have brought in a number of executives from Meta that have made a number of changes that erode the privacy of Firefox users. They can't ship any features that are hostile to Google directly because Google is their primary source of funding.

Safari is written by a company that has no vested interest in seeing the web be good enough to replace native apps because that company makes gobs of money being a middleman selling native apps.

Vivaldi is closed source.

LibreWolf, IronFox, WaterFox, Icecat, Cromite, Ungoogled Chromium are all similar in that they are drafting the work of the multinational corporations, but ultimately are volunteer efforts that are sustainable only as long as the volunteers continue to donate their time.

Brave is essentially trying to create a business doing what those privacy-focused browsers are doing. The challenge is they somehow need to monetize without compromising privacy, which has been a continual challenge for every browser manufacturer.

Again, I'm going to say it for the third time in this thread: the version of Brave that I use doesn't collect telemetry or serve ads because I paid for it. I can verify its behavior because it's also open source. Not a single person has responded to this very simple fact.

Yikes, didn't mean to trigger any defensiveness. Can't edit my comment, but mentally replace "bunch of performance issues" with what you said: "performance vs. security compromises".

Another example of how copyright makes preservation difficult.

I mean, it's obvious: copyright seeks to create a mechanism to compensate creators by creating scarcity. They succeeded.

It turns out archiving those creations effectively is in direct tension with that scarcity. So we made a system to incentivizes creation, and then destruction of the creation. This is visible with the "copyright cliff" in books, but also the incredible difficulty of legally keeping computer programs running over the decades. Copyright's restrictions on derivative works, coupled with the insanely long copyright term, create a sort of digital wasteland, with only those willing to go underground having great collections of old programs.

They are built around creating a user-centric browser, and need to monetize. They are actively moving to business models that are more privacy-focused (charging for the browser!) but no one talks about that.

Meanwhile, I'm not sure who you'd prefer. If we're pushing Firefox here, I'd strongly recommend folks compare and contrast Brave's policies with Firefox: https://www.mozilla.org/en-US/privacy/firefox/#how-is-your-d...

Take special care to read through the section called "To serve relevant content and advertising on Firefox New Tab".

I've argued this at some length elsewhere, so I don't feel compelled to do so here. You can drag up his $1,000 donation 20 years ago and the backlash it inspired at Mozilla that made him no longer effectively able to lead, but I don't see that as an indictment of his current efforts at all.

> If those who correct can’t or won’t explain why one is correct over the other, then it simply isn’t.

Or they aren't good at defending their position. Beware of people who are good at argument: they'll often seem to be right when they're not. Conversely, don't ignore someone just because they can't argue their point well. The point may have merit, even if they can't represent it as strongly as they should.

This is where the Hacker News guidelines about responding to the strongest possible interpretation of what somebody is saying come into play.

This isn't accurate. Mozilla stopped focusing on their core product, which was the browser. Brave has desperately been trying to monetize the creation of a privacy-focused browser for some time, much to the chagrin of folks on HN. I decided to pay for their open source offering, which doesn't have any of that.

The hate for Brave is some combination of people dredging up issues from very early in their existence and mischaracterizing them as often as not, and folks that insist that it's jammed full of crypto shit when it's very straightforward to either disable or remove it entirely by just using Brave Origin.

I have used all three of the browsers that you happen to mention aside from Firefox, and while I keep them in my back pocket, Brave is my daily driver. I don't tend to visit a huge variety of websites. I usually have JavaScript disabled when I do so, and I have other privacy preserving mechanisms in place besides just the browser I picked.

I'm happy for folks to trudge on with the Firefox engine, but I no longer see Mozilla as a good custodian of it. I do think the privacy-focused forks of it are good, and worthy of significant attention. On Android, I would also recommend both IronFox and Cromite.

Perhaps. I use the version you can pay for, and that doesn't have any of that.

Just as a counterpoint, I switched to Brave because Firefox no longer holds the technical or moral high ground. Brave is an open-source browser which makes it different from Chrome and Vivaldi. Brave is a privacy-focused browser. Brave is the only browser that is open source and also lets me pay for it so I can have a clean, uncluttered, fast experience. Brave is the only chromium-based browser that I'm aware of that continues to support Manifest V2 for UBlock Origin, which is critical for my use case.

Mozilla has been distracted by a million shiny things around them. They've been mismanaged almost criminally and they have changed their terms of service to be much less privacy-oriented than they used to be.

In my humble opinion, Brave gets far too much hate on this site and is actually the best browser made today in terms of open source availability, a focus on privacy and empowerment of the end user, while maximizing site compatibility.

Really? They're not using Scudo? The hardened_malloc used in GrapheneOS has a bunch of performance issues that were trade-offs against security. Every other major android distribution uses scudo as far as I know. Which OEMs are you thinking of?

Edit: Did some research after writing this? It appears that Samsung may be still using jemalloc, albeit a newer version than the one from the Android 11 days.

I've written about this before on HN, but it's pretty clear that once the app stores courted the hobby developers and convinced them to publish their apps on the stores to get the marketplace flywheel started (remember all the discussion about the app stores competing based on the number of apps available in each?), that the focus wasn't going to be on hobbyist developers at all because they don't bring in the most cash. By imposing ever higher restrictions on app developers in the name of security, they are raising the bar to such a degree that typically only funded corporate customers will have an incentive to go through all the effort, since they can literally pay someone to do so.

In essence, while the app stores used to be about hobbyist software to some degree, they have become a place created by big companies for big companies, much like Walmart or Target. That would be fine if customers could choose when they visited Walmart or Target, but instead, we've created a regime where you must buy everything from Walmart or Target.

This has made mobile development a wasteland for hobbyist-built and community-built software. I know some people cheer this as the removal of amateurs and crappy software from the marketplaces, but I would counter with the observation that commercial software often doesn't have my best interests at heart, and I would, many times, prefer a piece of community-built or amateur software instead. This gets harder to procure with each passing day.

So, I think developers are generally better-served leaving mobile development and instead focusing on web development, which is not gated in the same ways. While you won't have the same reach that you would get through the app stores, you do have much more control over the platform on which you're developing.

> Nothing is stopping people

The original game might be stopping people. It's not like games have a good track record of continuing to work for decades. DRM of all kinds poses a huge problem in this regard, but so does garden variety technological change. The GOG preservation team has quite a few videos about deeply technical work they needed to do to make e.g. Prince of Persia run properly on modern PCs. Sure I can just pirate a copy for PS2 and run it with PCSX2, but thats “piracy" (even though I still have a working original disc!).

So, no: there's a lot stopping people, especially copyright law, from keeping around working copies of their games.

> copyright expires

Not in your lifetime. 70 years after death if the author brings us out into centuries. Books written 60 years before my kids were born won't be out if copyright until they are in their 40s, for example.

> if you don't see the signs that authors are already being put out of work, then you need to get out of your bubble.

If it's so obvious, it shouldn't be so hard to articulate specifics. Maybe you can name just one author?

Being a writer was always a terrible profession for making money and having stable work. In my observations, AI is incapable of writing in the way a skilled author can -- what AI will replace is jobs working on small blurbs, promotional posters, and summaries. But those aren't really authors; I'm not sure if we care that the person employed to summarize novels at readers digest is now being replaced by an AI.

And that's why specifics matter. So much of the discourse is emotionally charged, and people imagine details that are not there. What I'd like to do is bring the details into the light.

So, does anybody reading this anywhere on this site have any information on any author that made money as a full-time writer that is now unemployed because their work has been replaced by AI? I'm not talking about the industry being upended. I'm a software engineer and my industry has been upended, but that has not resulted in job loss as much as it has changing what the profession does. Are authors in a different position than software engineers in this regard? Or are they truly just losing their jobs as they're mass replaced by AI? I haven't seen any evidence of the latter, which is why I'm asking.

I'm pointing out that the original title of the actual article is inaccurate and it would be helped by actual specifics. The actual plaintiffs in the lawsuit do not appear to have been put out of work.

Why would it be impossible? You're discounting the most probable case, which is that the AI writing is not a substitute for a skilled writer.

Edit: Turns out both Judge Alsup and Judge Chhabria agree that this is not obvious and would need to be defended. Here's Chhabria's statement on the evidence for market dilution:

> As for the potentially winning argument—that Meta has copied their works to create a product that will likely flood the market with similar works, causing market dilution—the plaintiffs barely give this issue lip service, and they present no evidence about how the current or expected outputs from Meta’s models would dilute the market for their own works.

When the plaintiffs don't even attempt to argue the point, that's pretty telling.

They're not skipping, they're keying on the exact same thing I am, which is whether or not authors have been put out of work because of AI, and to what degree. This is stated like it's a fact, but it is not a fact that is established, at least in my experience.

> even though they didn't say anything worth downvoting

Sometimes it's how you say things, not what you're saying.

You can get pretty far replacing your launcher with something that makes you type the name of the app you want to launch, and avoiding most apps, like FB, X, Amazon, Insta, etc.

The trillions are effective at forcing network effects, but anything outside that it pretty trivial to swap out and get a better experience.

I'm a bit surprised by the fatalism on display in this discussion. I get that it sounds like I'm victim-blaming, but you can only alter your own behavior, so assuming responsibility has merit.

This is a weird example. Are you addicted to your HSA app? It sounds like you are being intentional.

> And suggesting we abandon it as a comment on a social media platform is ironically self-unaware.

Gotta reach people where they are.

> Honestly only dumbasses wouldn't appreciate

I'm seeing this style of argument on the rise: "only an idiot would disagree". It reminds me of an old professor of mine that would use "proof by intimidation", where he would begin the proof with "As any fool can plainly see..."

It makes for terrible discussion.