No, it's not about translations (Source: I'm a NetHack fork developer who's somewhat involved in this DevTeam revival thing).

Translations of games like NetHack are inherently hard. You can't use the standard approaches as the program assembles the sentences out of several parts and usually, e.g. with gettext, you translate whole sentences. But here, we have dynamic sentences where this approach can't work.

For my German translation NetHack-De, I used internally latin-1 (so I could continue to use the char* strings) and for the output transformed the text into ASCII, latin-1, or utf-8 (depending on configuration).

The memory usage of UTF-8 is also at most char count multiplied by 4. The 5- and 6-byte sequences were declared invalid when Unicode was restricted to have no character above U+10FFFF.

Please, don't use the Wikia NetHack Wiki. It is outdated, ad-ridden, and has been abandoned by the community, but Wikia doesn't allow a wiki to be deleted.

The current NetHack wiki is at http://nethackwiki.com/ .

Pff, your loss.

UnNetHack: NetHack Improved!

I'm not asking for a flame war as much as trying to shame the dev team out of hiding!

Others have tried and failed (e.g. SporkHack was partly forked from Vanilla for this effect and this year several fork developer posted an April Fools' joke on RGRN about considering their fork the successor to NetHack and changing its name to "NetHack").

No reaction by the DevTeam. It's highly unlikely that the DevTeam will ever release another version (the development has already slowed down before 3.4.3), so you are stuck with the old version or forks or other roguelikes.

Vanilla NetHack hasn't had a release since 2003 but there have been several forks of it, one I did myself (look at my sig).

Considering the "far better roguelikes" that's something just asking for a flame war but I guess he thinks about ToME4 or Dungeon Crawl Stone Soup.

ToME4's root go back a long time, originally an Angband variant but the 4th version separated completely from that heritage and created vast amounts of original content that makes Skyrim look like a coffee-break activity.

Dungeon Crawl Stone Soup is sort of an Anti-NetHack, trying to avoid many of the design mistake NetHack had. Like the needs for spoilers, that different races play the same in the long run, grinding, or that the game doesn't stay challenging after a certain point.

DCSS and ToME4 are big games but in the last years there has been a trend to develop smaller roguelikes. Like DoomRL which is exactly what its title says or roguelikes for mobile devices like 100Rogues and POWDER.

I don't have high hopes that Braben can repeat the success of Elite. He's been talking about it for years, with small hints here and there but nothing has come from it.

So I'm mildly anxious that this time it will be for real.

But OTOH I don't expect it to be a big succes. In the best case it will be something along the lines of the recent XCOM remake and I would be quite content with such a game.

What you are saying, then, is that beginners should use a language that supports a REPL? Great! Now if we would just stop teaching Java to CS101 students...

REPL can be simulated on any language. I hope you're teaching your CS101 students more than just things that make sense in a REPL.

I don't think our philosophies differ that much. You didn't you would code and then immediately jump into debugger. You said "as soon as I feel like I understand what the code is supposed" so you already have thought about the code.

I've seen programmers trying to debug code by randomly and unsystematically putting breakpoints, hoping to getting a clue what the code does wrong, without even trying to think first. Shotgun debugging at its worst. If they don't have a clear model in their mind of what the code is supposed to do how can they deduce valid conclusions that can be proven or disproven?

Also: Memory protection leads to sloppy programmers. Bad habits should be punished.

Because of this sentence I'm not sure if your whole post is ironic or not. I'll treat it as non ironic.

Part of the learning process is not only in solving the exact thing you want to do. When you do the research needed to solve your problem you tend to get a lot of information that isn't related to the problem at hand but helps you understand the language and how to avoid related problems.

Yes, you tend to get a lot of information that is not related to the problem you're having but if you are a beginner but that doesn't mean that you also understand that information and keep it in mind for later reference. Because you just don't know the language and its problems yet!

If you're starting out in a new language (even if you already know programming in other languages) it is certainly not always obvious what code leads to unmaintainable code. Learning a language is always exploring unknown territory and exploring is much easier if you can do it in small steps and for that instant feedback is way better. You will always be using some sort of reference and tutorial for exploring it and if those tutorials recommend bad languages features or habits (and let's be honest, every language has these) then it's the fault of that learning material.

That may be, but tools made "for beginners" are bad tools. Not what you want when you're no longer a beginner. Then you want the beginner-stuff out of your way, so it doesn't waste your time.

I didn't say Eclipse is made "for beginners". IMO Eclipse works well enough OOTB even for advanced programmers, it stays out of your way most of the time.

But yes, if you want to become a really good programmer, you should expand your view beyond one single IDE. At least switching between Eclipse and Vim. :-)

Maybe I it's a bit of a rambling. But apparently the rest of my post is good enough to straighten out the weak intro. ;-)