You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I love your reflections. When you're done you should edit, make a 'process document' that includes the best parts, and your discussions. @masak's reflections (I think that Carl is the main vehicle when it comes to reflections in this project) and his discussions with @vendethiel and others.
My contention is that this would be valuable to others. Perhaps it could be included in some kind of 'history'/'process' section at the project homepage?
Another idea would be - ok, perhaps I'm stretching it now/again - to write 'a how to write a compiler' text and link to the different issues, so the reader can get a sense of (possible) problems, complications (issues) and so on. The main advantage would that the reader would see, 'meet' 'real' problems.
I think this would be particularly good for junior developers who don't grasp the whole process of writing a compiler, get a sense of 'real' issues (or how to put it). Often 'tutorials' or even whole books are way to 'easy'; you're provided with a solution, not the problems... and this is not always helpful. What not to do, is often equally important as what you should do (at least, I think, to the beginner programmer).
Who knows what lovely and cool spin-offs would emerge from such a procedure? It would be a truly honest project. And from honesty emerges the possibility to close in on truth, I would say.
Anyhow! I love reading your discussions and reflections.
The text was updated successfully, but these errors were encountered:
Re process document: thank you for the kind words. I think you will enjoy the planned fallout of the #401 documentation and also #415 (comment) — basically, my plan is to document for each language-extending module what forces went into shaping that feature the way it is, comparing and contrasting with alternatives.
Re "how to write a compiler" tutorial: I like the idea. As you're aware, there's a book in the very early stages. But your idea sounds like something smaller/earlier than that — maybe even something that can serve as an early version/prototype of the book. Being a fan of getting results early, I approve/agree. It will by necessity be a separate effort/project out-of-scope from 007, but of course we can link to the resulting post(s) from here.
Closing this issue due to its overlap with existing things.
I love your reflections. When you're done you should edit, make a 'process document' that includes the best parts, and your discussions. @masak's reflections (I think that Carl is the main vehicle when it comes to reflections in this project) and his discussions with @vendethiel and others.
My contention is that this would be valuable to others. Perhaps it could be included in some kind of 'history'/'process' section at the project homepage?
Another idea would be - ok, perhaps I'm stretching it now/again - to write 'a how to write a compiler' text and link to the different issues, so the reader can get a sense of (possible) problems, complications (issues) and so on. The main advantage would that the reader would see, 'meet' 'real' problems.
I think this would be particularly good for junior developers who don't grasp the whole process of writing a compiler, get a sense of 'real' issues (or how to put it). Often 'tutorials' or even whole books are way to 'easy'; you're provided with a solution, not the problems... and this is not always helpful. What not to do, is often equally important as what you should do (at least, I think, to the beginner programmer).
Who knows what lovely and cool spin-offs would emerge from such a procedure? It would be a truly honest project. And from honesty emerges the possibility to close in on truth, I would say.
Anyhow! I love reading your discussions and reflections.
The text was updated successfully, but these errors were encountered: