Agreed, but my point is that an abstraction should provide discernible gain of value, not merely roll a handful of operations of the same level into one.
1,196 karma · joined October 10, 2022
Senior Software Architect @ Austrian railways ÖBB
Blog: https://www.fabianzeindl.com/
(Other media: http://www.gnod.com)
The question is whether you are ok with the model naming someone who isn‘t listed, or failing to name someone who is listed.
Whenever they need to make two or three manual steps or configurations, they rather develop an abstraction layer where you just have to press one button to perform them.
Until that button needs to be accompanied by another manual action. Then they will again develop an abstraction which encapsulates the press of the button along with the augmenting action and is triggered by the pressing of a superior button.
Example: Docker->Kubernetes->Helm or any other tooling that uses YAML to write YAML.
I keep wondering why. All projects I ever saw need lines of code, nuts and bolts removed instead of added. My best libraries consist of a couple of thousand lines.
2. in my opinion, what all these tools are missing is the ability to add a manual „corrective“ layer, meaning to be able to adjust the automatically created output by moving and resizing boxes by a certain x/y amount.
That is an excellent remark.
As far as I know there‘s no other language which focussed from the get-go on parallel, distributed deployment, has processes as first class primitives, supports hot code swapping etc.
Erlang was invented at Ericsson, used for certain phone switches, it was tremendously successful at what it was built to do, and still it‘s not used anymore in their newer switches.
Why?
If the LLMs output is written in a way where I would edit at most 5% to be satisfied (which I did by the way), why shouldn’t I use it?
(Which, in my opinion has two reasons: 1. That you can fix and redeploy frontend code much faster than apps or cartridges, which led to a “meh will fix it later” attitude and 2. That JavaScript didn’t have a proper module system from the start)
I sometimes wonder if there could be an optimal number of microservices. As far as I know no one has connected issue data to the number of microservices before. Maybe there‘s an optimal number like „8“ which leads to lower number of bugs and faster resolution times.
I can imagine the discussions on the issue above with the outcome being either no icons, or icons on every item, but nothing in between.
How go about convincing a group of people about this?
The corresponding tasks always required disproportionately many high-level meetings and discussions for their actual simplicity.
I always attributed this directly to the complexity of the OpenSSL command and the fact that these tasks include hard deadlines because of certificate runtime.
The complex discussions which ultimately result in some simple commands mirror the OpenSSL CLI: you have to understand and articulate precisely what you need.
That being said I wonder some of the options (like "this file is a CER", PEM vs DER private key, etc.) couldn't be replaced by auto detection.
My point was not that reactions/feelings etc are switched by a signal being 1 or 0, just that sort-of-digital signals (line "on" or "off") are possible within the brain.
My idea about the instruction-set I was talking about earlier was not that it immediately switches behaviours on or off, but that it selects the next, slightly different or deeper "brain circuits / loops" to get activated and shift the whole apparatus towards finally acting out a certain behaviour.
But, as you point out, that is all unfounded speculation.
1. You ask someone you can trust for facts and opinions on topics, but you keep in mind that the answer might only be right in 90% of the cases. Also people tend to tell you if the are not sure.
2. For answers you need to rely on you ask people who are legally or professionally responsible if they give you wrong advice: doctors, lawyers, car mechanics, the police etc.
ChatGPT can‘t lose it‘s job if it informs you incorrectly.
That might be correct but as far as I know the brain has circuits that fire periodically with a certain frequency (what brain waves are about). If such frequencies steer some kind of gate than you do have a kind of digital logic with 1 being where the brain waves cross the zero line.
Meaning that while you are solving the rubik‘s cube, your brain has the capacity of switching into one of 2^10=1024 possible next states, which might fire nerve impulses for the body and also come with a new set of 1024 next states?
When focusing on the cube about 20 or 30 of the next states could be related to the cube, while the rest is concerned with bodily sensations, other thoughts etc.
What if learning something well meant that less states need to be reserved for the task/cube at hand because the decision tree going forward has become more efficiently packed, so more states are open for something else.
That might be the case, but the libraries seems to be more reusable!
> _for example_ by it's impact on bug resolution time
There are definitely ways to measure, for example by it's impact on bug resolution time. The following study looks at static code quality, but it could be applied to TDD:
This study is about static code quality, but could be applied to TDD as well. https://www.researchgate.net/publication/359129462_Code_Red_...
A fun question I tried a couple of times is asking it to give me a list with famous talks about a topic. Or a list of famous software engineers and the topics they work on.
A couple of names typically exist but many names and basically all talks are shamelessly made up.
Since http protocols and browsers change and especially google never ceases to push new optimizations, it is extremely difficult to keep up and do the right thing.