Re: why the pascal family of languages (Pascal, Ada, Modula-2,2,Oberon,

["Randy Brukardt" <randy@rrsoftware.com>]

"Dan'l Miller" <optikos@verizon.net> wrote in message 
news:5d9cd056-b36b-4448-be8d-a8bafbaa6d81@googlegroups.com...
>On Monday, July 21, 2014 7:01:20 PM UTC-5, Randy Brukardt wrote:
>> "Dan'l Miller" wrote in message:
>> > There actually does exist a GUI library written entirely in Ada for 
>> > Ada:
>> > Gnavi.
>>
>> You misspelled "Claw". ;-) GNavi exists mainly because the authors didn't
>> like our design for Claw. Which has been a long-standing problem with 
>> Ada --
>> too many choices never quite finished, rather than one library that most
>> people use and extend.
>
>Randy, you make it sound as though Gnavi is a fork of Claw or some other
>direct derivative work, by making the reference to misspelling (a rose by 
>any
>other name is just a sweet).

It's derivative in that it took the ideas that we had pioneered in Claw and 
reimplemented them. I didn't mean to imply that any of the actual code was 
used. Being first, we had the fun of debugging compilers (early Ada 95 
compilers were not up to the task when we started, particularly with 
finalization), and to a lesser extent, figuring out what doesn't work.

> Perhaps some of the developers of Gnavi were
>first Claw users, but it seems they simply designed a competitor rather 
>than a
>derivative work.
>
>
>  Or was now-EULA-licensed Claw open source at one point in the past?

The introductory version is open source now, and has been for a long time. 
That's not as obvious as it should be, mainly because I couldn't find the 
appropriate wording to put into the packaging. I planned to do that in the 
next release that's never happened.

The plan always was to open source the libraries (but not the tools). We 
didn't do it right away because it wasn't clear whether we had the right 
model, and we didn't want a lot of dependence on mistakes until we had more 
evidence (that is, programs using it) one way or another. (There also were 
the requirements of our ATIP/P contract, which needed be cleared before 
giving away anything.) Problems with early compilers pushed that back a 
number of years which effectively wrecked the plan. I had originally hoped 
to have what amounted to a standard Ada library for Windows, such that 
people could simply assume it is there and do cool stuff on top of it. That 
of course never happened in any sense.

At least a few parts of Claw ended up in the Ada standard: Ada.Directories 
and the various children of Ada.Calendar existed in some form in Claw 
(because these rather basic capabilities were missing from Ada 95), and the 
Claw versions formed the basis for the packages in the Standard (especially 
for Ada.Directories, which started as a copy of Claw.Directories with 
appropriate substitutions ["Claw" -> "Ada", etc.]).

                                     Randy.