                                                                         = Frequently Asked Questions about this web system

 * What is this web system and what is it called?    

  I am having trouble thinking of a name for this system, but
  I have been calling it 'webpad' or 'web notepad'. I am not 
  inspired at the moment to think of a name. 

  This system is a way of creating websites without using HTML
  and yet having freedom about what the content of the site is

 * What language is it written in?
 
  As you can probably tell, it is written in PHP

 * Do I need an SQL database for this to work?

  No, although it would be good if I could use one.

* How do I edit a page

  I still havent make an easy way to select a page to edit. But you can
  go to the control page, mentioned below.

  To edit the current page you can just click here [http://wagga-cwc.org.au/refurb/webpad/webpad.php?file=%2Fhome%2Fwaggacwc%2Fpublic_html%2Frefurb%2Fdocs%2Fsystem-faq.txt]

  You need to use the 'webpad' system which you can access through
  the 'control page' link:///refurb/webpad/control.php  This page allows
  you to find a file to edit. Not all files on the server are editable
  the the 'webpad' system. At least I certainly hope not. There is another
  system which does allow you to edit all the files but you will need
  a password for it

* Why did you write this?

  I was working in an enviroment in which I couldnt ftp to the 
  webserver which I was working on and for that reason I needed a 
  way to create web-sites through a web-browser interface. If you 
  have had a look at the 'zope' http://www.zope.org you will notice that
  it contains a system to do that. The only problem with that is that
  you need some fairly industrial strength priviledges on the web-server
  in order to set it up.

  I was trying to create a web-site for the Mission Australia
  community work center, and I got a very severe attack of overkill.
  Larry Wall, the original author of the perl language, has said that
  this is one of the standard programmers curses.


* What is a plain text file?

  Good question. You may have noticed, if you have been trawling through
  the various files which I have written, that I keep refering
  to this piece of terminology which does not seem particularly 
  valued in any other part of the non-unix world. I mean a file 
  which is encoded in one of the ISO standard character sets, such as
  ISO 8859-1 or perhaps
  a unicode standard like utf and which, in addition to this important
  characteristic, also contains no internal 'markup' structure, such as
  HTML or XML. Many people would call XML and HTML plain text files, but
  I dont. 

* What is markup?

  This is big question. Try searching the internet for the answer.

* Is this just another wiki system?

  Well, this is an accusation which I need to defend myself against. If you
  have had a look at the wiki systems available you will notice that there
  is a huge number, and it would seem very very odd to write another one.
  Fair enough. (please see Ward Cunninghams [http://www.c2.com] site for a
  good definition of what a wiki is. Since he invented the concept it 
  seems only fair).

  I dont really think that this system is a wiki system, even though 
  many of its features were inspired by wikis, especially the Oddmuse
  wiki [http://www.oddmuse.org]

  

* What features are you thinking of adding to the webpad system?
  
  a- Create configuration files to influence the appearance
     of the output html. 
   - Automatically generate style sheets for each document, if required.
   - allow copying of config options between documents
   - write output scripts for formats such as Unix Man, DocBook
     Tex Info, Latex, Groff, and so on ... This is because I have grandiose
     ideas about this system being a publishing system rather than
     just a way to create web-pages.
   - I would like to write an 'installer' script which would get all the 
     files required for the system from accross the internet and place them
     in the correct directories, as well as configuring the 'include' path
     which is at the top of each php script. This would also overcome many
     permissions problems with files and directories not being editable.
   - I need to template the HTML documents which are produced using the 
     function 'fnFillWebTemplate' link:///refurb/product-data/lib/
   - Automatically make text like 'warning: ' be formatted in red. Also,
     make references to keys on the keyboard get transformed into 
     images of those keys. So you would type "press enter" and the word
     enter would be transformed into an HTML image tag.
   


* When I edit documents I see strange codes at the bottom like [*faq*]. What does this mean?

  This tells the transformation engine which type of transformation
  system to use. We could call these transformations 'preprocessors'
  or filters etc. If a document is an faq it should be formatted
  in HTML a bit differently than if it is a journal or diary or 
  some other type of document. I will may get rid of these codes and
  allow this parameter to be set in a configuration file. 

* Why dont you just store all your documents in XML?    

  This is another very good question. The answer is partly ideological
  and partly practical. For example, I dont have a decent XML editor
  to use, even though I was supposedly attempting to write one for 
  the last year (2003) in Barcelona, but Barcelona was far too interesting
  a place to be writing XML editors in, so I didnt do much. The guy who
  has the company "findesign" http://www.findesign.org actually did a 
  lot of work on this but he wrote it in java and based it on 'netbeans'
  http://www.netbeans.org and in my opinion it was far too slow to be
  of any use. 

  The other thing is that I dont really like XML editors. I prefer to
  just type, preferably in an HTML textarea or in some obscure editor
  such as "vim" http://www.vim.org  So, this present system is going to
  try to be very xml friendly but it is not going to use xml as the 
  central data format. As Kernighan pointed out in a very old book on
  the Unix operating system (sorry no IBN, and by the way IBNs in the 
  current system dont create automatic links to 'amazon' 
  http://www.amazon.com like they do in a particularly interesting 'wiki'
  system called 'OddMuse' http://www.oddmuse.org 

* This thing seems to be getting a bit too complex. How will you maintain it?

  Actually, I dont know. The transformations are really a little difficult
  to debug now, and when I add templating functionality to these 
  documents, its going to get even harder to debug. I am going to 
  use a 'preprocessor' approach to configuring the appearance of the 
  final HTML according to options which can be set by the user in a
  configuration file using a webpage interface.

  Also, the tree of function dependencies is now getting difficult to
  keep track of, since in some cases there is about 4 levels of function 
  calls. 



[*faq*]                                                                                                        