|
| 1 | +Implementation Status (work in progress) |
| 2 | +========================================== |
| 3 | + |
| 4 | +* Grammar: 99.99% done but new ideas come up every day... |
| 5 | +* Shell: implementation sorted out & working on the basic modules... |
| 6 | +* Language: The C/C++ Programming Language |
| 7 | + |
| 8 | +This is not a dead project. ;-) |
| 9 | + |
| 10 | +Planned features: |
| 11 | + |
| 12 | +- I/O Redirection (non-blocking pipelines all the way) |
| 13 | +- Background Jobs (sure) |
| 14 | +- Deferred Commands (could be more dangerous with this shell language) |
| 15 | +- Subshells (could be useful without openning a new shell) |
| 16 | +- Null Expansion Error (empty expansion stings; still has nullable expansion) |
| 17 | +- Variable Scopes (default to local scope; no implicit block scopes) |
| 18 | +- Namespaces (for module management) |
| 19 | +- Ternary Logic (needed for the 21st century) |
| 20 | +- Globbing (fail glob, null glob, fail/null regexp) |
| 21 | +- Cartesian Products (just like bash's {a,b,c}{1,2,3}) |
| 22 | +- Multi-line Strings (no here documents; echo "<multiple lines>" | cat) |
| 23 | +- String Interpolations (any handy shell script needs this feature) |
| 24 | +- Array, Unordered Map, Ordered Map (containers for text) |
| 25 | +- Perl Compatible Regular Expressions (power without perl) |
| 26 | +- Arbitrary-precision Arithmetic (calculator included) |
| 27 | +- Command Auto-completion (ease the typing) |
| 28 | +- Event Listeners/Triggers (spooky at a distance) |
| 29 | + |
| 30 | + |
| 31 | +Why another shell program? |
| 32 | +========================================== |
| 33 | + |
| 34 | +First of all, it is not a replacement of sh/bash/zsh/... |
| 35 | +It is not intended to rule the POSIX world. |
| 36 | +It is created because many (legacy) shells just f*ck my work style. |
| 37 | + |
| 38 | + * bash: arcane syntax, RTFM first |
| 39 | + * zsh: bloated syntax, must RTFM because too magical |
| 40 | + * rc: nice implementation, poor user interface and polluted env |
| 41 | + * fish: attractive but going bad, ugly details waiting inside |
| 42 | + * ????: why bother with another true (and ugly) programming language? |
| 43 | + |
| 44 | +Good-enough things are unbelievably great excuses to kill personal styles. |
| 45 | +Now I gotta try my own implementation. |
| 46 | + |
| 47 | +Goals: |
| 48 | + |
| 49 | + * ease of typing (well, mainly for US keyboards) |
| 50 | + * comprehensible syntax |
| 51 | + * no undefined behaviors |
| 52 | + * centralized UI configuration |
| 53 | + * finely commented source code |
| 54 | + * portable across unix-y systems |
| 55 | + |
| 56 | +Non-Goals: |
| 57 | + |
| 58 | + * posix-compliant |
| 59 | + * pipe structured data |
| 60 | + * interoperable with bash or other shells |
| 61 | + * one scripting language to rule the world |
| 62 | + * provide an escape-meta-alt-control-shift ide |
| 63 | + * rich multi-language/locale support (defects: os/encoding/font/terminal) |
| 64 | + |
| 65 | +In summary, the shell should ease your typing and provide a simple scripting |
| 66 | +language that is easy to learn and master, thus more time for fun projects. |
| 67 | + |
| 68 | + |
| 69 | +Scripting Language |
| 70 | +========================================== |
| 71 | + |
| 72 | +The shell scripts will use a new scripting language. |
| 73 | +The actual look of it is left as an exercise for other shell developers. |
| 74 | +(May there be a *Unix Shell Haters' Handbook*?) |
| 75 | + |
| 76 | +Here is a checklist: |
| 77 | + |
| 78 | +Programming Language Checklist |
| 79 | +by Colin McMillen, Jason Reed, and Elly Jones. |
| 80 | + |
| 81 | +You appear to be advocating a new: |
| 82 | +[ ] functional [*] imperative [ ] object-oriented [*] procedural [ ] stack-based |
| 83 | +[ ] "multi-paradigm" [ ] lazy [ ] eager [ ] statically-typed [*] dynamically-typed |
| 84 | +[ ] pure [ ] impure [ ] non-hygienic [ ] visual [*] beginner-friendly |
| 85 | +[ ] non-programmer-friendly [ ] completely incomprehensible |
| 86 | +programming language. Your language will not work. Here is why it will not work. |
| 87 | + |
| 88 | +You appear to believe that: |
| 89 | +[*] Syntax is what makes programming difficult |
| 90 | +[*] Garbage collection is free [*] Computers have infinite memory |
| 91 | +[*] Nobody really needs: |
| 92 | + [ ] concurrency [ ] a REPL [ ] debugger support [*] IDE support [ ] I/O |
| 93 | + [ ] to interact with code not written in your language |
| 94 | +[ ] The entire world speaks 7-bit ASCII |
| 95 | +[ ] Scaling up to large software projects will be easy |
| 96 | +[*] Convincing programmers to adopt a new language will be easy |
| 97 | +[ ] Convincing programmers to adopt a language-specific IDE will be easy |
| 98 | +[ ] Programmers love writing lots of boilerplate |
| 99 | +[ ] Specifying behaviors as "undefined" means that programmers won't rely on them |
| 100 | +[*] "Spooky action at a distance" makes programming more fun |
| 101 | + |
| 102 | +Unfortunately, your language (has/lacks): |
| 103 | +[*] comprehensible syntax [?] semicolons [_] significant whitespace [_] macros |
| 104 | +[_] implicit type conversion [_] explicit casting [_] type inference |
| 105 | +[_] goto [_] exceptions [_] closures [?] tail recursion [_] coroutines |
| 106 | +[_] reflection [_] subtyping [_] multiple inheritance [_] operator overloading |
| 107 | +[_] algebraic datatypes [_] recursive types [_] polymorphic types |
| 108 | +[_] covariant array typing [_] monads [_] dependent types |
| 109 | +[_] infix operators [_] nested comments [*] multi-line strings [*] regexes |
| 110 | +[*] call-by-value [_] call-by-name [_] call-by-reference [_] call-cc |
| 111 | + |
| 112 | +The following philosophical objections apply: |
| 113 | +[*] Programmers should not need to understand category theory to write "Hello, World!" |
| 114 | +[*] Programmers should not develop RSI from writing "Hello, World!" |
| 115 | +[ ] The most significant program written in your language is its own compiler |
| 116 | +[*] The most significant program written in your language isn't even its own compiler |
| 117 | +[ ] No language spec |
| 118 | +[ ] "The implementation is the spec" |
| 119 | + [ ] The implementation is closed-source [ ] covered by patents [ ] not owned by you |
| 120 | +[ ] Your type system is unsound [ ] Your language cannot be unambiguously parsed |
| 121 | + [ ] a proof of same is attached |
| 122 | + [ ] invoking this proof crashes the compiler |
| 123 | +[ ] The name of your language makes it impossible to find on Google |
| 124 | +[*] Interpreted languages will never be as fast as C |
| 125 | +[ ] Compiled languages will never be "extensible" |
| 126 | +[*] Writing a compiler that understands English is AI-complete |
| 127 | +[ ] Your language relies on an optimization which has never been shown possible |
| 128 | +[ ] There are less than 100 programmers on Earth smart enough to use your language |
| 129 | +[ ] ____________________________ takes exponential time |
| 130 | +[ ] ____________________________ is known to be undecidable |
| 131 | + |
| 132 | +Your implementation has the following flaws: |
| 133 | +[*] CPUs do not work that way |
| 134 | +[*] RAM does not work that way |
| 135 | +[ ] VMs do not work that way |
| 136 | +[ ] Compilers do not work that way |
| 137 | +[ ] Compilers cannot work that way |
| 138 | +[ ] Shift-reduce conflicts in parsing seem to be resolved using rand() |
| 139 | +[ ] You require the compiler to be present at runtime |
| 140 | +[ ] You require the language runtime to be present at compile-time |
| 141 | +[ ] Your compiler errors are completely inscrutable |
| 142 | +[*] Dangerous behavior is only a warning |
| 143 | +[ ] The compiler crashes if you look at it funny |
| 144 | +[*] The VM crashes if you look at it funny |
| 145 | +[*] You don't seem to understand basic optimization techniques |
| 146 | +[*] You don't seem to understand basic systems programming |
| 147 | +[*] You don't seem to understand pointers |
| 148 | +[*] You don't seem to understand functions |
| 149 | + |
| 150 | +Additionally, your marketing has the following problems: |
| 151 | +[*] Unsupported claims of increased productivity |
| 152 | +[*] Unsupported claims of greater "ease of use" |
| 153 | +[ ] Obviously rigged benchmarks |
| 154 | + [ ] Graphics, simulation, or crypto benchmarks where your code just calls |
| 155 | + handwritten assembly through your FFI |
| 156 | + [ ] String-processing benchmarks where you just call PCRE |
| 157 | + [ ] Matrix-math benchmarks where you just call BLAS |
| 158 | +[*] Noone really believes that your language is faster than: |
| 159 | + [*] assembly [*] C [*] FORTRAN [*] Java [*] Ruby [*] Prolog |
| 160 | +[*] Rejection of orthodox programming-language theory without justification |
| 161 | +[ ] Rejection of orthodox systems programming without justification |
| 162 | +[ ] Rejection of orthodox algorithmic theory without justification |
| 163 | +[ ] Rejection of basic computer science without justification |
| 164 | + |
| 165 | +Taking the wider ecosystem into account, I would like to note that: |
| 166 | +[ ] Your complex sample code would be one line in: _______________________ |
| 167 | +[*] We already have an unsafe imperative language |
| 168 | +[ ] We already have a safe imperative OO language |
| 169 | +[ ] We already have a safe statically-typed eager functional language |
| 170 | +[*] You have reinvented Lisp but worse |
| 171 | +[ ] You have reinvented Javascript but worse |
| 172 | +[ ] You have reinvented Java but worse |
| 173 | +[ ] You have reinvented C++ but worse |
| 174 | +[ ] You have reinvented PHP but worse |
| 175 | +[ ] You have reinvented PHP better, but that's still no justification |
| 176 | +[ ] You have reinvented Brainfuck but non-ironically |
| 177 | + |
| 178 | +In conclusion, this is what I think of you: |
| 179 | +[ ] You have some interesting ideas, but this won't fly. |
| 180 | +[*] This is a bad language, and you should feel bad for inventing it. |
| 181 | +[ ] Programming in this language is an adequate punishment for inventing it. |
0 commit comments