Does anyone have any experience or results with using system prompts that intentionally make LLMs more annoying to encourage better habits?I believe that consistently using an LLM for creative or solution building tasks will: - degrade critical thinking skills - degrade knowledge - encourage apathy - degrade knowledge acquisition skills - trades short term benefit for heavy long term lossI...
Imagine your basic Excel spreadsheet -> generating document files, but add:
- Other sources like SQL queries
- User form (e.g. "Generate documents for Client Category [?]")
- Chaining sources in order like SQL queries with parameters based on the user form
- Split at multiple points (5 records in a csv, 4 records in a sql result = 20 generated documents)
- Full Jinja2 templating with field substitution but also if/for blocks that works nicely with .docx files
- PDF output
- output file names using the same templating: "/BusinessDrive/{{ client_id }}/Invoice - {{ invoice_id}}.pdf"
All saved in reproducible workflows (for example if you need to process a .csv file you receive each morning)
Hi there, this post is introducing AutoDocument, a free and open-source document generating web app that connects spreadsheets, databases and user forms into documents such as Microsoft Word and PDFs. It's based on fantastic open sources libraries like https://github.com/elapouya/python-docx-template and headless LibreOffice.Mail Merge is a pain because it:- Only converts from Excel to Word-...
Hi HN,Since leaving my job I've created a prototype of a project that would have helped me immensely in that job.It is essentially a self-hosted document creator where you can chain all kinds of sources together (CSV/excel, SQL queries, API calls, user forms etc) where the output of one can be the input of the next and create 1 or more documents (Word, PDF, HTML, all kinds of outputs) using a...
If the database knew what queries were running ahead of time and have them compiled you could easily defend against SQL injection because the SQL isn't being interpreted as it's run, and it would be possible to run analysis over the query to see if there are any indexes or implementations missing that could speed it up. Something like a stored procedure's variable substitution but not...
I'm just responding to that quote from your last post.
> When you wear particular clothing, it communicates which group you identify with or want to associate with.
Agree to a point, but I don't think "Boring" is generally a group someone would try to identify/associate with (I'm assuming you're using boring as a derogatory term here). I also disagree that outgoingness is synonymous with how interesting a person is (in fact I'm sure you'd agree there are many instances of the contrary).
And outside of being "good" or "bad" at non-verbal communication you're forgetting another important direction: people who do not care.
Take you for example - I find you very interesting because you have strong opinions that differ to mine - It would have been a shame had I immediately categorised you as boring given the sunglasses you were able to afford or if you didn't care about sunglasses beyond their function.
Found this comment rubbed me the wrong way - seems insane to me to judge the boringness of a person by the price of their sunglasses
Reminds of this xkcd: https://xkcd.com/435/
The beauty of programming is abstraction - you write a function to do one thing, do it well, and then you can abstract that concept away and reuse that abstraction. Although it's only an abstraction - a container is "like" a lightweight VM, and you can use it like that until it doesn't act like a lightweight VM. In which case, you have to dabble one layer deeper just to understand enough to know why the abstraction of your layer isn't a perfect analogy.
In which case, if you're looking for an expert in layer n, then basic knowledge of layer n - 1 might be a decent proxy for expertise