Commit dd64f86e authored by Rob Pike's avatar Rob Pike

another bite-sized checkpoint on the language design FAQ

R=rsc,iant
DELTA=87  (54 added, 2 deleted, 31 changed)
OCL=35058
CL=35061
parent 24bfaaf0
...@@ -29,7 +29,7 @@ ...@@ -29,7 +29,7 @@
<li class="navhead">Related Guides</li> <li class="navhead">Related Guides</li>
<li><a href="go_tutorial.html">Tutorial</a></li> <li><a href="go_tutorial.html">Tutorial</a></li>
<li><a href="go_spec.html">Language Specification</a></li> <li><a href="go_spec.html">Language Specification</a></li>
<li><a href="go_lang_faq.html">FAQ</a></li> <li><a href="go_faq.html">FAQ</a></li>
<li class="blank">&nbsp;</li> <li class="blank">&nbsp;</li>
<li class="navhead">Other Resources</li> <li class="navhead">Other Resources</li>
<li><a href="./">Go Docs</a></li> <li><a href="./">Go Docs</a></li>
...@@ -46,32 +46,37 @@ ...@@ -46,32 +46,37 @@
Do not delete this <div>. --> Do not delete this <div>. -->
<div id="nav"></div> <div id="nav"></div>
<h2 id="origins">Origins</h2>
<h2 id="creating_a_new_language"> <h3 id="creating_a_new_language">
Why are you creating a new language?</h2> Why are you creating a new language?</h3>
<p> <p>
TODO TODO
</p> </p>
<h2 id="history"> <h3 id="history">
What is the history of the project?</h2> What is the history of the project?</h3>
<p> <p>
TODO TODO
</p> </p>
<h2 id="ancestors"> <h3 id="ancestors">
What are Go's ancestors?</h2> What are Go's ancestors?</h3>
<p> <p>
Go is in the C family, but also borrows some ideas from CSP-inspired Go is mostly in the C family (basic syntax),
languages such as Newsqueak and Limbo. The interface idea may be with significant input from the Pascal/Modula/Oberon
related to other languages but was designed in isolation; ditto family (declarations, packages),
packages. In every respect the language was designed by thinking plus it borrows some ideas from languages
inspired by Tony Hoare's CSP,
such as Newsqueak and Limbo (concurrency).
However, it is a new language across the board.
In every respect the language was designed by thinking
about what programmers do and how to make programming, at least the about what programmers do and how to make programming, at least the
kind of programming we do, more effective, which means more fun. kind of programming we do, more effective, which means more fun.
</p> </p>
<h2 id="protagonists"> <h3 id="protagonists">
Who are the protagonists?</h2> Who are the protagonists?</h3>
<p> <p>
Robert Griesemer, Rob Pike and Ken Thompson laid out the goals and Robert Griesemer, Rob Pike and Ken Thompson laid out the goals and
original specification of the language. Ian Taylor read the draft original specification of the language. Ian Taylor read the draft
...@@ -80,8 +85,10 @@ Cox joined later and helped move the language and libraries from ...@@ -80,8 +85,10 @@ Cox joined later and helped move the language and libraries from
prototype to reality. prototype to reality.
</p> </p>
<h2 id="different_syntax"> <h2 id="change_from_c">Changes from C</h2>
Why is the syntax so different from C?</h2>
<h3 id="different_syntax">
Why is the syntax so different from C?</h3>
<p> <p>
Other than declaration syntax, the differences are not major and stem Other than declaration syntax, the differences are not major and stem
from two desires. First, the syntax should feel light, without too from two desires. First, the syntax should feel light, without too
...@@ -94,8 +101,8 @@ descendants are notoriously difficult in this regard but it's not hard ...@@ -94,8 +101,8 @@ descendants are notoriously difficult in this regard but it's not hard
to fix things up. to fix things up.
</p> </p>
<h2 id="declarations_backwards"> <h3 id="declarations_backwards">
Why are declarations backwards?</h2> Why are declarations backwards?</h3>
<p> <p>
They're only backwards if you're used to C. In C, the notion is that a They're only backwards if you're used to C. In C, the notion is that a
variable is declared like an expression denoting its type, which is a variable is declared like an expression denoting its type, which is a
...@@ -109,7 +116,7 @@ the declaration ...@@ -109,7 +116,7 @@ the declaration
int* a, b; int* a, b;
</pre> </pre>
<p> <p>
declares a to be a pointer but not b; in Go declares <code>a</code> to be a pointer but not <code>b</code>; in Go
</p> </p>
<pre> <pre>
var a, b *int; var a, b *int;
...@@ -132,19 +139,19 @@ is not just the expression grammar; keywords such as <code>func</code> ...@@ -132,19 +139,19 @@ is not just the expression grammar; keywords such as <code>func</code>
and <code>chan</code> keep things clear. and <code>chan</code> keep things clear.
</p> </p>
<h2 id="no_pointer_arithmetic"> <h3 id="no_pointer_arithmetic">
Why is there no pointer arithmetic?</h2> Why is there no pointer arithmetic?</h3>
<p> <p>
Safety. Without pointer arithmetic it's possible to create a Safety. Without pointer arithmetic it's possible to create a
language that can never derive an illegal address that succeeds language that can never derive an illegal address that succeeds
incorrectly. Compiler and hardware technology has advanced to the incorrectly. Compiler and hardware technology have advanced to the
point where a loop using array indices can be as efficient as a loop point where a loop using array indices can be as efficient as a loop
using pointer arithmetic. Also, the lack of pointer arithmetic can using pointer arithmetic. Also, the lack of pointer arithmetic can
simplify the implementation of the garbage collector. simplify the implementation of the garbage collector.
</p> </p>
<h2 id="inc_dec"> <h3 id="inc_dec">
Why are <code>++</code> and <code>--</code> statements and not expressions? And why postfix, not prefix?</h2> Why are <code>++</code> and <code>--</code> statements and not expressions? And why postfix, not prefix?</h3>
<p> <p>
Without pointer arithmetic, the convenience value of pre- and postfix Without pointer arithmetic, the convenience value of pre- and postfix
increment operators drops. By removing them from the expression increment operators drops. By removing them from the expression
...@@ -158,25 +165,68 @@ with the STL, part of a language whose name contains, ironically, a ...@@ -158,25 +165,68 @@ with the STL, part of a language whose name contains, ironically, a
postfix increment. postfix increment.
</p> </p>
<h2 id="absent_features">Absent features</h2>
<h3 id="generics">
Why does Go not have generic types?</h3>
<p>
Generics may well come at some point. We don't feel an urgency for
them, although we understand some programmers do.
</p>
<p>
Generics are convenient but they come at a cost in
complexity in the type system and run-time. We haven't yet found a
design that gives value proportionate to the complexity, although we
continue to think about it. Meanwhile, Go's built-in maps and slices,
plus the ability to use the empty interface to construct containers
(with explicit unboxing) mean in many cases it is possible to write
code that does what generics would enable, if less smoothly.
</p>
<p>
This remains an open issue.
</p>
<h3 id="exceptions">
Why does Go not have exceptions?</h3>
<p>
Exceptions are a similar story. A number of designs for exceptions
have been proposed but each adds significant complexity to the
language and run-time. By their very nature, they span functions and
perhaps even goroutines; they have wide-ranging implications. There
is also concern about the effect exceptions would have on the
libraries. They are, by definition, exceptional yet experience with
other languages that support them show they have profound effect on
library and interface definition. It would be nice to find a design
that allows them to be truly exceptional without encouraging common
errors to turn into special control flow requiring every programmer to
compensate.
</p>
<p>
Like generics, exceptions remain an open issue.
</p>
<h3 id="assertions">
Why does Go not have assertions?</h3>
<p>
This is answered in the general <a href="go_faq.html#Where_is_assert">FAQ</a>.
</p>
<h2 id="TODO"> <h3 id="TODO">
TODO</h2> TODO</h3>
<p>TODO:</p> <p>TODO:</p>
<pre> <pre>
Why does Go not have: Why does Go not have:
- assertions - macros?
- exceptions - conditional compilation?
- generic types
What do you have planned? What do you have planned?
- variant types? - variant types?
explain: explain:
package designa package design
slices slices
oo separate from storage (abstraction vs. implementation) oo separate from storage (abstraction vs. implementation)
goroutines
why garbage collection? why garbage collection?
...@@ -184,6 +234,7 @@ why garbage collection? ...@@ -184,6 +234,7 @@ why garbage collection?
no data in interfaces no data in interfaces
concurrency questions: concurrency questions:
goroutine design
why aren't maps atomic why aren't maps atomic
why csp why csp
...@@ -198,6 +249,7 @@ oo questions ...@@ -198,6 +249,7 @@ oo questions
why no automatic numeric conversions? why no automatic numeric conversions?
make vs new make vs new
Why do maps only work on builtin types?
</pre> </pre>
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment