In 30 places, it's also very illegal to do business with vendors who ransom your data, if you're in finance, i.e. an entity covered by the Digital Operational Resilience Act; NIS2 (27 places) doesn't spell it out but also requires business continuity planning. Natural persons in the EU+EEA also retain a right to data portability under GDPR and there are data access/portability provisions in the EU Data ACT and DMA. Many legal frameworks require the covered entity to be 'in control' of vendors and data. Proactive legalese allowing the vendor to ransom your data is not quite in line with that requirement; in many sane jurisdictions such clauses would be found unenforceable.
As far as i understand, this is not +-300ms. It is 300ms vs. 10 sec or something. That is a huge difference. I personally find the time to wait for these larger models a limiting factor. It’s also probably a resource waste for fairly simple task like this. (Compared to the general function approximation of the llms)
But I honestly feel like the task of smartly applying edits falls somewhat within traditional coding tasks. What about it is so difficult it could not be done with a smart diffing algorithm?
you misunderstood. its 300ms just for the apply model, the model that takes your coding models output (eg sonnet) and figures out where the code should be changed in the file. Cursor has its own, and claude uses a different technique with strings as well. So its 10sec vs 10sec +300ms using your analogy
For someone not heavy in this space. I use GH copilot at work. I might switch to cursor. I am not into the details of the tools just care does it help me or not. For us it might be worth having an easier to understand value proposition.
It may take a bit if explaining and that's OK. But the big question is as someone doing my enterprise microservice who isn't heavy into AI why do I switch to you.
What are they protecting against? Honestly. LLMs should probably have an age limit, and then, if you are above, you should be adult enough to understand what this is and how it can be used.
To me, it seems like they only protect against bad press
Yes, it is indeed to mitigate bad press. Unfortunately, the discussion about AI is so ridiculous, that it is often considered newsworthy when a product generates something funky for a person with large enough Twitter audience. Nobody wants to answer the questions about why their LLM generated it and how they will prevent it in the future.
Kind of serious question. Do we have any alternatives to html? If not, why? It’s essentially all html. Yes, browser will render svg/pdf/md and so on, but as far as I can tell, it’s not what I consider "real web" (links to other documents, support for styling, shared resources, scripting, and so on ).
I would have loved for there to be a json based format, or perhaps yaml, as an alternative to the xml- based stuff we have today.
>links to other documents, support for styling, shared resources, scripting, and so on
SVG supports all those things. SVG's use element can do exactly what the OP is talking about. You can put HTML in SVG and SVG in HTML. SVG is just markup exactly like HTML. SVG supports JavaScript and stylesheets. SVG can put text on a curve and other stylish things that boring boxy HTML cannot accomplish.
Read it. Ch 11 shows you how use can import chunks of other SVG. Ch 12 is JavaScript. You can inline it in your html and style it with CSS just like any of the other tags in html. Once you know SVG, you can hand generate more compact images than anything created by something like Inkscape which just uses path elements for everything. For example, you can do a UPC barcode with a single SVG line element using a wide stroke and a stroke-dasharray. See chapter 8.
I’m a noob to this subject. How can a build be non-reproducible? By that, I mean, what part of the build process could return non-deterministic output? Are people putting timestamps into the build and stuff like that?
Timestamps, timestamps, absolute paths (i.e., differences between building /src versus /home/Cort3z/source), timestamps, file inode numbering ("for file in directory" defaults to inode order rather than alphabetical order in many languages, and that means it's effectively pseudorandom), more timestamps, using random data in your build process (e.g., embedding a generated private key, or signing something), timestamps, and accidental nondeterminism within the compiler.
By far the most prevalent source of nondeterminism is timestamps, especially since timestamps crop up in file formats you don't expect (e.g., running gzip stuffs a timestamp in its output for who knows what reason). After that, it's the two big filesystem issues (absolute paths and directory iteration nondeterminism), and then it's basically a long tail of individual issues that affect but one or two packages.
reply