Re: Windows Printing

[Lennart Borgman]

Eli Zaretskii wrote:
> > Eli and I (with a lot of others involved) had a very long discussion 
> > about this on the help-gnu-emacs mailing list:
> >
> >    http://lists.gnu.org/archive/html/help-gnu-emacs/2006-01/msg00242.html
> >
> > I think the tests I did clearly showed that everything does not work as 
> > the node above says.
> >     
>
> That's really not an accurate summary of what was written in that
> discussion.  A more accurate summary, IMHO, would be that _sometimes_,
> in rare situations with strangely configured network printers, neither
> the default printer port detection that is part of the Emacs startup
> on Windows, nor any of the tips and tricks in the manual, succeed in
> getting Emacs to print to that printer.  (Other printers on the same
> system did work, IIRC.)
>   
Sorry, I did not want to exaggarate. But are you really sure that what 
you write here is correct? It is true that I detected one printer on the 
network that worked the way that Windows Printing in Info describes. The 
others did not.

However that is not my point. I just want to stress that the information 
in Info should be augmented. Otherwise we might waste some peoples time.

>   
> > My conclusion was that the state of the printer is 
> > not reset when accessed in the way Emacs does it on w32. Getting the 
> > printer port is done in a standard way (I mean according to MS 
> > documentation) but sending the data is not supposed to be done as Emacs 
> > does it (if I understand this correctly, see the thread).
> >     
>
> Sorry, I disagree with this conclusion.  The MS docs do not say
> explicitly that ``sending data is not supposed to be done as Emacs
> does it'', and the simple fact is that the way we do it in Emacs works
> on the vast majority of Windows systems.
>   
I am very sure about the conclusion that the printer is not reset when 
accessed from Emacs. What exactly do you disagree upon?

You might be right that this works for the majority of Windows system - 
I am not sure. But as above how can you be sure? Perhaps it is a bit 
like "all people I know vote for that"? Maybe some people gives up on 
Emacs because they find the information inaccurate? (I know some do, 
because I once did.)

We might be reading the MS documentation differently. But I tried to 
point to some parts of the manual (in the thread on the mailing list) 
that made me come to the (tentary) conclusion that printing on MS 
Windows should be done in a different way than Emacs currently does it. 
The supported way (as I read the MS docs) is through the use the GDI 
interface (which I know very little about in practice). Grabbing the 
name of the printer port is a different issue.

> The use of Ghostscript is covered by the manual, in the node you
> mentioned.  As for htmlize.el, I refuse to recommend Emacs users to
> use Explorer or Notepad to print on Windows.  Sorry.
>   
A little bit unfair. I do not recommend Notepad or Explorer. I use 
htmlize.el + Firefox ;-)

Then perhaps a link to http://EmacsWiki.org/ could be in the Info node? 
And perhaps it could be pointed out that the only recommended way that 
works for everyone on MS Windows is through the use of Ghostscript+GSview?

>   
> > What should we do about this node?
> >     
>
> If you or someone else can debug the problem you've been having and
> find out how to work around it, I'll be happy to augment the manual.
> Otherwise, I don't see what we can or should do.
>   
Both you and I did a lot of work around this. I finally said that I can 
not see any way to get further. In the situation where I found the 
problems I simply can not in a general way do more.

What can be done is sending reset sequencies to the printers, but those 
are printer specific and I would really not recommend that. Maybe it 
would help some people to mention this possibility in the manual, but it 
would in my opinion look a bit ridiculous nowadays and therefore it 
would not be in favour of Emacs or GNU.