• 7 Posts
  • 356 Comments
Joined 3 years ago
cake
Cake day: July 26th, 2023

help-circle

  • I get the point and idea behind the existence of the “most people understand…” argument. A hot dog isn’t actually made from dog. But when it comes to truth vs lies… if you aren’t an entertainer, you just shouldn’t be allowed to lie. And if you are an entertainer, it needs to be beyond obvious in you marketing and such that you are an entertainer. If you channel is called anything news, you are not an entertainer. Google search can be entertaining, but almost no one would say it is entertainment. So there is just no justification for it lieing with impunity.













  • One class seems too few. 75% of jobs in the US use computers regularly. Even the plumber shows up with a tablet and such. On top of that, what ever was the point in making us write the essay twice, rough draft, the final… and have to rewrite the whole damn thing if we made a mistake. When it comes to writing, computers are where it is done.
    Math… yeah, pencil and paper, calculator for the high level stuff. History/social studies… videos and articles are just easier to distribute via computer. Though initial presentation with follow up commentary is ideal. I think computers are overused in school, but 1 class is too far in the other direction.





  • I don’t favor forced adoption… but your second sentence just isn’t true of individual people. It is ture of the industry in general. But individual people tend to stick with what they know and shy away from new things until something forces them out of that state. Usually it is someone else who tried a new tool and is suddenly able to do things easier or what not. It only takes a few power users to turn the tide. So forcing shouldn’t be needed. Just enabling those power users.


  • hm. Well I tend to work with smaller companies that have less people overall. So we need a lot of automation to enable the devs to be more productive. We have fully automated cicd pipeline where everything in the pipeline can be run locally. Most of the scripts support the pipeline allowing quick and easy reproduction of anything that might not work in gitlab. We have scripts to authenticate, validate, build, deploy, unit test, integration test, performance test, lint, e2e test, vuln scan, terraform, teardown, updating tickets, generating release notes… It’s a very transparent system to the developers. And it works pretty well, which is why they don’t need to have all the scripts memorized. They just don’t need to run them much. But when the pipeline fails in a way they don’t understand, access to all this allows them to self-serve finding out why. I know that plenty of places just make the devs do a lot of this manually. And since they are doing it often, it’s easy to remember and a required part of the job. But humans make mistakes, or cut corners and such which leads to bugs getting out into production. With working automation the devs don’t have to do these things manually very often which means they don’t need to memorize all of them, and less opportunity for bugs to slip through that could have been caught. The fact that they don’t have our scripts memorized is actually a sign of success for the automation. I have always been of the opinion that the more we can automate the mundane tasks away, the better. The people I have worked with prefer spending their time designing and coding, not memorizing mundane processes. Even with all our automation, our devs are always hoping for more.


  • I’m not so sure about that. Real time voice recognition is still pretty poor in my opinion. Chat takes that out of the way. So the failure of voice doesn’t mean chat can’t work. And think of how often you use grep. It’s pretty effective. AI can be a more advanced form or grep. And most importantly, we can tune where it looks and such. So like an easier to configure search engine, or really several that can be tuned for the specific search. If a search engine in the form of an AI skill can be highly specific, it can be better then a general search engine. Now put like 10 or 15 of those together with specifics to how your specific team/company works. And toss configuring the AI on top of that to know which one to use for what. Feels like we could make it very good. But of course, it isn’t going to work out of the box. We have to take ownership of it and make it work. Then give it to the whole team…