pvo50555

That is the spot where the shoe-shine boy was? Is there a marker for where Daguerre's studio was?

That's it - that's the smoking gun.

Local, small-batch RAM! Top fermenting Get it at your local farmer's market!

I loved Thief 1/2, these games were way ahead of their time and defined the genre for years to come.

I am consistently shocked at how bad Apple Notes is. I like the features it has and I like the iCloud sync between devices... I use it daily. But it's so damn buggy. It's a notes app! Why do new lines on a moderately-long note cause half the text to disappear until I hit the carriage return a few times? I also wish it simply converted copied text to plaintext by default rather than trying to preserve formatting.

You are not my son.

For my purposes, all of the data was also available in HTML format, so the OCR wasn't a problem. I think the issue is the RAG pipeline doesn't take the entire corpus of knowledge into its context when making a response, but uses an index to find one or more relevant documents that it believes are relevant, then uses that small subset as part of the input.

I'm not sure there's a way to get what a lot of people want RAG to be without actually training the model on all of your data, so they can "chat with it" similar to how you can ask ChatGPT about random facts about almost any publicly available information. But I'm not an expert.

What differentiates this from Open WebUI? How did you design the RAG pipeline?

I had a project in the past where I had hundreds of PDF / HTML files of industry safety and fatality reports which I was hoping to simply "throw in" and use with Open WebUI, but I found it wasn't effective at this even in RAG mode. I wanted to ask it questions like "How many fatalities occurred in 2020 that involved heavy machinery?", but it wasn't able to provide such broad aggregate data.

What does "throw everything into the model" entail in your context?

How much data are you able to feed into the model in a single prompt and on what hardware, if I may ask?

> Regarding iOS, they have made some progress in adopting webpush, but it does not seem possible to push for mobile applications with it yet.

This is a key point right at the very end. It has always been much easier to implement push notifications to Android devices, even when only Web Push is implemented. Google has been the main force behind the adoption of PWAs as Native apps on devices, even by immediately showing a "Install this app?" popup when navigating to a PWA-enabled website/web app.

iOS permits "Installing to home screen" but there is no popup shown natively, nor any indication from the OS that the website you're visiting is a PWA. The user has to go through several steps to add the app to his device.

Once it is added, it's possible for the device to receive web push notifications and show them as native push notifications (IIRC), even with the app is closed, however the reliability is inconsistent. Sometimes you get them, sometimes you don't. At least that was my experience still about 6 months ago when I tried implementing web push with PWA for iOS.

I gave up trying to go the pure PWA route and instead wrapped my web app with PWABuilder into a native app, and went through the whole process of submitting to the App Store, getting APN, registering the web app in Firebase, then sending the push notifications to the native app via Firebase Messaging. Works great then, but it's a lot more work than keeping everything simply as a web app and using web push.

When did you manage to get that? I tried signing up for the ARM server several times but it always says it's unavailable at the moment in my region (East - Toronto) and I should try a different region... however you can't switch regions after you set up your account. I talked to their support and they said it's unavailable for the foreseeable future... choose a different configuration.

Are people still able to provision those instances in other regions? If so, which one(s)? I'm just using the legacy free one for now, the 2GHz AMD with 0.5GB RAM.

You know those guys? My friend and I used to come to tears to these during the old DubToons days. It's the only reason I know what a Middlesborough accent is. We still quote them!

I can't believe this has 1587 points (already). Has no one here grown up with those Magic Eye books?

Is it actually able to generate all the files and directory structure, etc? Or did the author just take all the responses he got from multi-shot prompting and eventually assemble it into the final product? I believe I used LlamaCoder some time ago to build the scaffolding of an app, but I'm not sure about the state of that project now.