First of all thank you guys for the replies and suggestions.

@FND
>Out of curiosity: Why rewrite everything from scratch rather than using 
>(or at least investigating) existing libraries? 

Perhaps I have misstated my intentions here. I am definitely looking into 
existing libraries most notably libcurl and libhubbub for parsing the HTML. 
I haven't made any definite decisions yet as to implementation rather I am 
still brainstorming on the best way to achieve all of my desired results.

The problems which arise seem to revolve around security. I am in love with 
the ease of use, versatility and small learning curve which TW provides, 
but if I am to bring these ideas even closer to the core of a local system 
then I must consider security above all else.

The idea of building these things from scratch stem from the fact that 
browsers are now and will always be tightening security (and rightly so, 
because the web can be a hostile place and local filesystem access is a 
scary thing). With security getting tighter and the core of TW consisting 
of web technologies, I need a way to create, render, modify and delete 
content all while maintaining the integrity of the data itself and the 
integrity of the local filesystem.

My first solution was to re-write just TW in C and handle the rendering of 
content myself. This would allow me to have one TW which acts as a system's 
operating environment and since it would only handle local data, security 
would be left up to common sense on which tiddlers and TW's you import, and 
once imported we can use the OS's permission system (in Linux: rwxrw-rw-) 
to restrict access to each tiddler based on your permissions or your 
group's permissions.

This approach has some problems with it which I would need to solve before 
I start coding, but honestly I thought I might've insulted the entire TWDev 
community by continuing to pursue a rewrite of the core especially at a 
time when the creator of TW himself is also pursuing a rewrite for HTML5 
(excellent work so far by the way). I would hate to fork the community, 
especially when his plans are actually way more useful than mine would be 
in the end.

@Chris Dent

First off, great work on TiddlyWeb! 

It is true that a lot of your ideas and implementations solve most of the 
problems I would run into, and in fact if I ever do start coding this thing 
I would be borrowing heavily from your work (all credit will be given of 
course even if I only borrow the ideas and not the code).

>have you considered forking them to get at the core presentational and 
>javascript functionality they provide, and then adding in your own 
>support for tiddlers-on-disk? 

I have looked into this especially with respect to the gecko rendering 
engine since TW has always worked better with FF. This has not been 
discounted (and might in fact be my solution), but there would need to be a 
lot of work there as well in porting the ideas to a local-files-only 
paradigm.

The main idea of mine which seems to be creating the most problems is that 
I think in developing the operating environment tiddlers and TW files which 
are local should default to be opened by a separate program from other HTML 
files. This separate program should allow for direct access to the local 
filesystem in accordance to the users permissions.

alright I've started rambling and repeating myself, so I'm going to gather 
my thoughts and ideas and try to decide on a course of action. If I 
actually made sense here and anyone has anything to add please don't 
hesitate to post a reply.

Thank you guys so much for your time and consideration.


-- 
You received this message because you are subscribed to the Google Groups 
"TiddlyWikiDev" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/tiddlywikidev/-/BeQ1N368eP8J.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/tiddlywikidev?hl=en.

Reply via email to