[pracjourn-forum] For our error collection

[Peter Flynn]

On Fri, 2005-09-09 at 14:51, Karl Berry wrote:
>     that is what defines a program which is allowed to call itself "TeX".
> 
> This hardly matters any more.  Web2c-TeX has not been "TeX" for anything
> except (plain) "tex" for many years.  

But that is precisely what the TRIP test defines. Plain TeX.

> The fact that Knuth wrote certain
> messages back in 1978/1982 should hardly stop us from making them better
> now; I'm sure he himself would not want that.

On the contrary, this is exactly what he has requested: make whatever 
changes you like, but don't call the result "TeX", call it something 
else, because "TeX" refers to the program *he* wrote.

So we are free to change anything, but it would have to be called Splosh
or Wurd or PageBroker or something...

> However, I fear that Peter (Flynn) is right.  No matter how much better
> error messages would be desirable and useful, the barrier to changing
> anything in core TeX is astronomically high.  It is a real problem.

No, there is no barrier at all (except the daunting complexity of
reprogramming something written as tightly as TeX). We just have to 
rename it.

> Peter (Flynn), I don't buy the "front end" argument, any more than I buy
> the "front end" argument for why certain things should not be in CSS or
> HTML (according to Phil Taylor, anyway).  

I'm not familiar with Phil's arguments on this, not having met him for
many years. Doubtless a few beers at a UK TeX meeting would fix that :-)

> The reality is that (a) most,
> or at least many, people just run (La)TeX, not anything else, and (b) no
> front ends that I know of actually do anything useful to change error
> messages.

My argument exactly: because most users now are using LaTeX, there is no
reason why front ends shouldn't be able to do this. I guess I'll have to
write the proof-of-concept myself though.

///Peter