Part of my job is working on trying to make these models productive for the large corporation I work for. It's a lot of throwing tomatoes at a wall and to a degree I see the issue he is talking about output seemingly having a certain ceiling.
At the same time in no part of his post is any code snippet or anything to latch on to of "the model performed poorly here when it should have done this" - this style of criticism seems to be a pattern of most of these "the LLMs will never work" style posts on blogs and twitter.
They obviously can perform better than autocomplete and in my own day to day development build out huge portions of a codebase that I would have expected a junior or midlevel engineer to perform at.
How are we really supposed to grasp their actual capabilities when no one will actually cite specifically what mistakes they are making.
Your work (and gyroscope/stethscope/other aggregators) has inspired me to also start my own work at creating a centralized aggregator to ingest all of my data.
Great work! The only thing is that this needs to get updated based on the changes of the new 0.9.9 release. Specifically you do not need to create a custom dispose function, remove now will remove listeners bound to the view.
Cool project - one of the major pain points with Yeoman was trying to get templating working. Cool to see a project which takes this approach from the beginning.
Congratulations to the Grove.io team! Its great that this product is able to keep going.
I have been following Grove.io pretty closely. This is a subject that is fairly interesting to me and compelled me to look at some of the different open source options[0], as they do exist for teams that are able to run their own chat. I decided to write my own chat client with a fellow HNer called Subway[1], since I could not find a good open source web client that persisted a connection while a user was offline. After I started I found another large project that more or less accomplishes the same thing called Alice[2] written in Perl. Both these clients have more or less the same goal to persist your connection while you are offline. For those interested in a paid alternative to grove there is also IRCCloud[3] which accomplishes the same thing.
Grove inspired me to make an open source alternative: Subway https://github.com/thedjpetersen/subway . And while it is far from perfect it is a start towards opening up IRC to teams.
Congratulations to the Django team. This is a big step towards to the adoption of python 3. The more big libraries that are ported the sooner we will see wide spread adoption.
At the same time in no part of his post is any code snippet or anything to latch on to of "the model performed poorly here when it should have done this" - this style of criticism seems to be a pattern of most of these "the LLMs will never work" style posts on blogs and twitter.
They obviously can perform better than autocomplete and in my own day to day development build out huge portions of a codebase that I would have expected a junior or midlevel engineer to perform at.
How are we really supposed to grasp their actual capabilities when no one will actually cite specifically what mistakes they are making.