nalaginrut

I never met TTN physically, but I'm familiar with him in cyberspace from his code. I didn’t realize he was a private person in real life. From his work, I always felt he was someone deeply passionate, someone who expressed his creativity wholeheartedly through code.

He was one of the pioneers of GNU Guile. It was a lonely and thorny road. Guile was an underappreciated project with a small community, limited tooling for exploration, and very few reference projects to learn from. Only true lovers of the craft were willing to contribute under such conditions.

TTN had spent his precious life time to help to make the foundation of GNU Guile community. His works and skills in the code inspired me a lot in the early time.

I’ve heard that he also contributed to other free software projects, though what I’ve written here is what I know firsthand.

It says, the internet has its own memory. I hope that by writing this to remember what he did and the quiet impact he had.

It seems somebody doesn't like Pu'er, it's not bad, unforgettable aroma. Whether you like it or not, its scent will flood the entire room, even if you tip it down the toilet.

nice try!

Maybe it's time to face cyberpunk seriously, in life, not in game.

Oh terrible!

If someday this GizVault becomes a super sensitive keyword that no one can post it on Hacker News, that means they win anyway. Everyone heard about it.

But it is still so weird why someone is so fear about it?

It's a pity, I hope they can criticize the content, isn't there any neat girls/boys come to roast them a little? It seems they're winning.

Don't let them win, come on girls and boys! Beat their theory seriously!

If that's the case, I wonder whether languages with steep learning curves still have a place, now that GC offers a more straightforward approach that requires less mental overhead.

You mean functionality as in functional programming? I wonder if it aligns with Unix philosophy. But that's an interesting insight.

He was a great Guile hacker, and his work inspired me a lot. RIP

Agreed, we actually have a Lua-5.2 frontend underway. We may officially support Lua in the future, but for now, we need to support full r7rs in Scheme.

Yes, it's a neglected topic, but it's the topic that LambdaChip wants to try.

It's a really interesting and challenging question. I've written a boostraping compiler, and I enjoyed the developement. It's a strange and sleepless process.

Well, back to the topic, maybe we can challenge it for LambdaChip someday in the future. ;-)

For the version issue, I have to say when I started to research Lua, there's only 5.2 BNF I can find on the internet. So I decided to try it from 5.2. Of course, I'm willing to catch up with the newer version.

For ISSUE-1. Yes, I know it may break the compatibility of some existing code. The problem is why people reference a variable named 'self' but not expect it to be the current table? People may named a var as 'self' to hold other value. I think that is the problem. 'self' is actually a reserved word in Lua, but Lua permit it to be used for other value. I don't think it's good. For example, some Scheme implementations allow users redefine primitives, you may redefine '+' with '-'. It's interesting, but it's bad when you want to build serious program. Another point is mentioned in the article that people could easier find the error when they encountered an exception related to 'self'.

Anyway, my point is that a good Lua programmer should avoid to use a.b.c.foo() if one wants to get the current table, no matter how the function is defined (colon or point). But people always do wrong thing (include me), so why not provide a mechanism to let them avoid this problem? Fixed ISSUE-1 could let people get rid of this problem, even you make the mistake to use point reference on 'self', you can get correct value if you defined the function with colon.

I'm very sorry, the problem is raised by folks comments, it's all my bad, I should build a better blog. Now I closed the comments, if you have any comments please leave it here.

I've no any doubt people will understand how good it is for various kind of coroutines usage when they come to Guile. What we need is more libs and projects for reducing workload in practices.

That seems not as glue code, although I agree Python win in preliminary scientific computing.

Yes it'unstable, it's Beta. Mysql/poatgre/sqlite3 are expected to work. So if you have problem, that's bug, please report it.

You may need to modify blog.scm to specify your DB passwd to make it work. I'll write a tutorial to show how to play it like Rails/Django.

Artanis is still young, but anyway, we are on the way.

When you put all things into a blackbox and pray for the promising result, the whole methodology goes to a dead end because it's not revealing the truth but voodoo magic to self-cheating. Nowadays, people just put big data into a blackbox to make "AI". We need the AI system to tell us why, not how. Or it's not AI to help you, you made yourself a Lord to give you order. And you never have chance to ask it why.

Symbolic AI could be a chance, because it keeps necessary meta info to reveal reasonable things. But how to optimize it properly will be a problem.