Posted by: Anonymous Coward
on April 20, 2006 07:50 PM
would you not rather leave it at Postscript and fix CUPS?
I don't consider a PDF based printing backend and an improved CUPS to be mutually exclusive. Let's take a look at PDF again: PDF is one of the prefered document interexcange formats. Most programs that are interested in printing are also able to produce PDF files, and you can get virtual printer drivers that produce PDFs, meaning that it is generally easy to produce PDFs. Many printers, especially the more professional ones, can read PDF directly, and if not, their driver or the printing subsystem could convert it as has been done with PostScript; the printer down the hall here even has a build in webserver that allows me to upload PDF files and have the printed that way. PDF is as mentioned in the article, a safe exchange format. PDF is a better format for print than PostScript.
So PDF can generally be considered superior to PostScript, and if PostScript were to be replaced, PDF would be a good candidate.
But is it worth it to rip out PostScript from the printing subsystems and replace it with PDF? Well, you mention that CUPS is a mess, but could it be that retrofitting it with PDF could provoke the amount refactoring that would make CUPS streamlined? If we consider that the printers talk PDF natively, and the user space applications talk PDF natively, wouldn't it make everything cleaner if the printing subsystem also talked PDF natively? The risc of bugs are generally lower with a clean architecture, and also consider that whenever you convert data from one format to another, a dataloss is generally to be expected and keeping everything in PDF will prevent this too.
Learn about Linux
Download Linux
Get Linux help
Re:Congratulations
Posted by: Anonymous Coward on April 20, 2006 07:50 PMI don't consider a PDF based printing backend and an improved CUPS to be mutually exclusive.
Let's take a look at PDF again:
PDF is one of the prefered document interexcange formats.
Most programs that are interested in printing are also able to produce PDF files, and you can get virtual printer drivers that produce PDFs, meaning that it is generally easy to produce PDFs.
Many printers, especially the more professional ones, can read PDF directly, and if not, their driver or the printing subsystem could convert it as has been done with PostScript; the printer down the hall here even has a build in webserver that allows me to upload PDF files and have the printed that way.
PDF is as mentioned in the article, a safe exchange format.
PDF is a better format for print than PostScript.
So PDF can generally be considered superior to PostScript, and if PostScript were to be replaced, PDF would be a good candidate.
But is it worth it to rip out PostScript from the printing subsystems and replace it with PDF?
Well, you mention that CUPS is a mess, but could it be that retrofitting it with PDF could provoke the amount refactoring that would make CUPS streamlined?
If we consider that the printers talk PDF natively, and the user space applications talk PDF natively, wouldn't it make everything cleaner if the printing subsystem also talked PDF natively?
The risc of bugs are generally lower with a clean architecture, and also consider that whenever you convert data from one format to another, a dataloss is generally to be expected and keeping everything in PDF will prevent this too.
So all in all, PDF has my backing.
#