Hacker Newsnew | past | comments | ask | show | jobs | submit | thomasmarcelis's commentslogin

Also something about subscriptions vs pay-for-usage. I feel the need to use all my weekly tokens or I'm wasting and I bet they would never get this kind of usage out of me if AI ended up being same price per token.


I always buy software/assets/dev tools for my hobbies (like CAD, music production, game dev) instead of paying subscriptions, even if that would very likely be way, way cheaper and would give me access to really cool tools. I don’t want to feel bad not using something and I know that’s the case with a subscription


What are you using it for? It has been completely subpar compared to any other LLM for me.

It's so bad at understanding your intentions.


I've been using it for setting up infra and projects on GCP and its been great. I use cursor for coding, but that isnt as helpful responding outside the IDE on cloud config. I have no GCP experience and I was able to get to a working application very quickly with Gemini. The GCP docs are outdated, often conflicting, but the Gemini experience was excellent.


Yes, daily. It's extremely useful, superior to internal search while combining the internal knowledge base with ChatGPT's


Probably same PR company or law firm


>>In our final scenario, the NCC Group consultant got booked on a scenario-based assessment:

>>“Pretend you have compromised a developer’s laptop.”

Most companies will fail right here. Especially outside of the tech world security hygiene with developer's laptops is very bad from what I have seen.


I also can't help but wonder this. I'm certain people at google search want to provide the best quality search results and do this with integrity. But at some point in the business hierarchy you are at a level where people set objectives for both these departments ( search & ads ) and are trying to optimise for things like total revenue/profit.


Thats close to our use case. We use it to sync a small amount of configuration data across a small amount of servers (2-50, depending client).

RQlite was perfect as our software is an addon for a legacy platform. We needed an easy low-access way of installing a distributed datastore.


Cool. Did you use read-only nodes, by any chance?

https://github.com/rqlite/rqlite/blob/master/DOC/READ_ONLY_N...



Everything makes sense now. 0.3% is an order of magnitude lower than the fees paid by US merchants.


They had API access but extracting and using the data in the way they did was against the terms of use IIRC


Essentially. The root failure mode of FB's API solution is that it's an API for accessing private data and their only protection against clients abusing that access was a terms of service.

Back in the mid-'00s, we started implementing something against FB's API and I was more than a bit shocked to discover that their solution for preventing someone from building a parallel network of friend relationships out of the available dataset was "Please don't do that, and anyway if you try to do it at scale we'll probably detect the bandwidth consumption and shut you down." Not great.


Sure looks like it. Part one, two and three are written by three different "Guest authors", who are freelance technical writers.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: