Walter,
I am becoming increasingly anxious about the stability and dependability of
DMD and D, with respect to the protection attributes and storage-class
attributes. It appears that either the documentation, DMD or both are
wrong. There has been a flurry of recent discussion about these attributes,
and there is a history of such discussions too, but there is no resolution
yet. You are the only one that can resolve the issues that have been
presented but you are silent on this, and that is causing my anxiety.

I felt that I didn't understand what you really wanted for D so I attempted your documentation yet again. I ended up rewriting and reorganizing the documentation to help me see your specification for D attributes, plus I worked though many sample programs to test out DMD against the specification. It would appear that DMD and its documentation are in serious conflict with each other.

I submit my revised documentation here as a possible starting point for a clearer picture and resolution of the issues.

You can view my docs online

  http://www.users.bigpond.com/ddparnell/attr.html

and the files that they were generated from

  http://www.users.bigpond.com/ddparnell/attr.d
  http://www.users.bigpond.com/ddparnell/style.css

These files are now in the public domain.

I am expecting that you will read and correct this documentation, and pass back your findings to either me or the newsgroup.

-- 
Derek
(skype: derek.j.parnell)
Melbourne, Australia
"Down with mediocrity!"
27/06/2006 4:56:29 PM

Walter? Who is Walter? A kid like play hide-to-seek?

Sorry,just get up. :)

"Derek Parnell" <derek@nomail.afraid.org> ??????:1xf0m4urq2q9n.1xwrr4voyhwwk$.dlg@40tude.net...
> Walter,
> I am becoming increasingly anxious about the stability and dependability
> of
> DMD and D, with respect to the protection attributes and storage-class
> attributes. It appears that either the documentation, DMD or both are
> wrong. There has been a flurry of recent discussion about these
> attributes,
> and there is a history of such discussions too, but there is no resolution
> yet. You are the only one that can resolve the issues that have been
> presented but you are silent on this, and that is causing my anxiety.
>
> I felt that I didn't understand what you really wanted for D so I
> attempted
> your documentation yet again. I ended up rewriting and reorganizing the
> documentation to help me see your specification for D attributes, plus I
> worked though many sample programs to test out DMD against the
> specification. It would appear that DMD and its documentation are in
> serious conflict with each other.
>
> I submit my revised documentation here as a possible starting point for a clearer picture and resolution of the issues.
>
> You can view my docs online
>
>  http://www.users.bigpond.com/ddparnell/attr.html
>
> and the files that they were generated from
>
>  http://www.users.bigpond.com/ddparnell/attr.d
>  http://www.users.bigpond.com/ddparnell/style.css
>
> These files are now in the public domain.
>
> I am expecting that you will read and correct this documentation, and pass back your findings to either me or the newsgroup.
>
> -- 
> Derek
> (skype: derek.j.parnell)
> Melbourne, Australia
> "Down with mediocrity!"
> 27/06/2006 4:56:29 PM
> 

Maybe, Walter find a good begin, but has no ability to reach the end. Maybe,what we want is a d like language controled by myself, the programer.

"Derek Parnell" <derek@nomail.afraid.org> ??????:1xf0m4urq2q9n.1xwrr4voyhwwk$.dlg@40tude.net...
> Walter,
> I am becoming increasingly anxious about the stability and dependability
> of
> DMD and D, with respect to the protection attributes and storage-class
> attributes. It appears that either the documentation, DMD or both are
> wrong. There has been a flurry of recent discussion about these
> attributes,
> and there is a history of such discussions too, but there is no resolution
> yet. You are the only one that can resolve the issues that have been
> presented but you are silent on this, and that is causing my anxiety.
>
> I felt that I didn't understand what you really wanted for D so I
> attempted
> your documentation yet again. I ended up rewriting and reorganizing the
> documentation to help me see your specification for D attributes, plus I
> worked though many sample programs to test out DMD against the
> specification. It would appear that DMD and its documentation are in
> serious conflict with each other.
>
> I submit my revised documentation here as a possible starting point for a clearer picture and resolution of the issues.
>
> You can view my docs online
>
>  http://www.users.bigpond.com/ddparnell/attr.html
>
> and the files that they were generated from
>
>  http://www.users.bigpond.com/ddparnell/attr.d
>  http://www.users.bigpond.com/ddparnell/style.css
>
> These files are now in the public domain.
>
> I am expecting that you will read and correct this documentation, and pass back your findings to either me or the newsgroup.
>
> -- 
> Derek
> (skype: derek.j.parnell)
> Melbourne, Australia
> "Down with mediocrity!"
> 27/06/2006 4:56:29 PM
> 

In article <e7sref$tgg$1@digitaldaemon.com>, Boris Wang says...
>
>Maybe, Walter find a good begin, but has no ability to reach the end. Maybe,what we want is a d like language controled by myself, the programer.

Walter has the ability, maybe just not the time...

This type of thing ("community development of the language and compiler") has been discussed before quite a bit and I think we, as D programmers, are getting the best of all worlds. A BDFL (Benevalent Dictator for Life) in Walter who has implemented world-class compilers before, knows a lot about language design and who is also doing the majority of the work, _while_ we still get some say in the design and implementation of the language, compiler and libraries.

Generally, if you come up with a really good idea that fits in w/ D semantics and goals and has some community support, Walter may at least consider implementing it, time permitting.

We all have our pet issues with the language design, but I think the whole is more than the sum of the parts as-is right now.

There's only so much you can fit into one language v1.0 spec. after all <g>

I think the best thing we as D users could do right now is find bugs in the compiler and/or where the spec. doesn't match the implementation and report them.

It looks like that's what Derek had in mind.

>
>"Derek Parnell" <derek@nomail.afraid.org> ??????:1xf0m4urq2q9n.1xwrr4voyhwwk$.dlg@40tude.net...
>> Walter,
>> I am becoming increasingly anxious about the stability and dependability
>> of
>> DMD and D, with respect to the protection attributes and storage-class
>> attributes. It appears that either the documentation, DMD or both are
>> wrong. There has been a flurry of recent discussion about these
>> attributes,
>> and there is a history of such discussions too, but there is no resolution
>> yet. You are the only one that can resolve the issues that have been
>> presented but you are silent on this, and that is causing my anxiety.
>>
>> I felt that I didn't understand what you really wanted for D so I
>> attempted
>> your documentation yet again. I ended up rewriting and reorganizing the
>> documentation to help me see your specification for D attributes, plus I
>> worked though many sample programs to test out DMD against the
>> specification. It would appear that DMD and its documentation are in
>> serious conflict with each other.
>>
>> I submit my revised documentation here as a possible starting point for a clearer picture and resolution of the issues.
>>
>> You can view my docs online
>>
>>  http://www.users.bigpond.com/ddparnell/attr.html
>>
>> and the files that they were generated from
>>
>>  http://www.users.bigpond.com/ddparnell/attr.d
>>  http://www.users.bigpond.com/ddparnell/style.css
>>
>> These files are now in the public domain.
>>
>> I am expecting that you will read and correct this documentation, and pass back your findings to either me or the newsgroup.
>>
>> -- 
>> Derek
>> (skype: derek.j.parnell)
>> Melbourne, Australia
>> "Down with mediocrity!"
>> 27/06/2006 4:56:29 PM
>> 
>
>