Re: TSR blacklist

[Jo Even Skarstein <joska@nuts.edu>]

On Tue, 9 Jun 1998, Martin-Eric Racine wrote:

> I'm really getting tired of utilities that try to cram everything in;
> currently, I have several applications that all duplicate each other
> on one function or the other, all because some basic things were not
> included in the basic OS.
> 
> Stuff like:
> 
>    *  case-sensitive fileselector
>    *  mouse accelerator
>    *  functional desktop (eg: Thing)
>    *  desktop clock
>    *  windowed dialog calls
>    *  scrollbox and listbox calls
>    *  fontselector calls
>    *  buffered, non-blocking, multi-tasking friendly GDOS printing
>    *  window topper
>    
> should have been integrated into the AES features by standard. Since

So you're expecting 1995-features in a OS designed in 1984?

> it wasn't done, we end up with a mixture of TSR and ACC that each use
> their own custom AES library, all conflicting with one another.

Partially correct. There are some badly written TSR's out there, but
you'll soon learn to avoid them. That some people (hi Juhani!) insists on
using them despite their shortcomings is their problem...

AES-libraries isn't a problem at all, although I admit that some
extensions to the AES would have been nice. The problem is that nobody
really use them. How many apps do you see that use MultiTOS-like toolbars,
which has been around for several years?

> IMHO, it's about time N.AES and MagiC people sit down and draft specs
> on what new system calls should be added, then release a Compendium II
> to document the new functions.

I'm afraid that means that ASH will push their wdialog-extension, which is
a bad solution if you ask me... I'm not saying that they're not willing to
listen, but instead of proposing standards they just develop something and
the rest of the Atari-community doesn't know about it until it's too late
to discuss them. Examples: Weird 3D in MagiC-AES, XFS-interface, shared
libraries, windowed dialog extensions...

> Peter, this is not a criticism of Clocky. Several people have praised
> it and it's probably very good. ;-) 

It actually obliterates most of your problems if you're the kind of person
that likes to have your auto-folder full of little widgets.

> Already, NEWSie tries to have its own window topper; Thing, xUFSL and
> Wdialog each have their own fontselector, etc. 

The point is that there are good and established protocols available, but
lots of programmers ignore/doesn't know about them. I constantly see new
apps that has a built-in, crappy font-selector instead of calling xufsl or
using FONTSELECT. Many new applications doesn't understand drag'n drop or
even the AV-protocol. Lots of them doesn't even conform to ancient,
well-knows GEM-standards.

> There are too many incompatible and duplicated functions and protocols. 
> It's got to stop.  Let's have the MagiC, Milan and N.AES people sit down
> together and write a new standard AES library. It's badly needed. NOW.

No it's not. The last thing we need is yet another protocol or API. People
should start using what we already have (AV, OLGA, FONTSELECT, DHST,
BubbleGEM...) instead of insisting on making 100% self-contained apps that
works on floppy-based 1.00 machines. 

I've been thinking about putting together a web-site with descriptions of
all the modern, powerful, *existing* protocols we have, perhaps this is a
good time to start...

Basically, I disagree with your first argument (designing a new,
multi-purpose AES-extension), but agree with your second (collecting
existing docs into one single, consistent document.) What happened to
Thomas Much's "TOS-Book"?


/*
** Jo Even Skarstein   http://www.stud.ntnu.no/~josk/
**
**   beer - maria mckee - atari falcon - babylon 5
*/