PandoraWiki -> Evopedia


Darkknight512

Active Member
Joined
Sep 7, 2010
Messages
637
Age
30
Location
Canada
I think it might be a good idea to convert the entire PandoraWiki to an Evopedia archive and either make it easily downloadable or include it with the firmware if it is not too big and use that as the main manual.
 
Never used evopedia, but indeed, having the wiki embedded is a good idea.
 
It's not a bad idea, just not a practical one. Firstly, the wiki will almost certainly be too big, and you really don't want to be putting *anything* on the NAND that isn't absolutely necessary.


Also, the wiki is only barely keeping up to date with things as it is (not that Esn et al. aren't doing a great job, though), so I doubt we want to be putting more pressure on the maintainers by changing to EvoPedia. Even if we did manage to move without too much trouble, the user would then need to keep downloading new versions of the wiki as it is constantly being updated, or have a dynamically generated firmware using the latest version.


And after all of that, this will cause a lot of people to think what they have on their NAND is the be-all and end-all of advice, and not check the more recent online version.
 
Last edited by a moderator:
compressed raw text is obviously going to be huge! like.. the man pages are 700GB of information and couldn't be included in the firmware by default.
 
compressed raw text is obviously going to be huge! like.. the man pages are 700GB of information and couldn't be included in the firmware by default.
I'm assuming that was was sarcastic, if it wasn't just ignore this post.

  1. Compressed raw text is admittedly not big, but a lot of the pages have important images (just look at the hardware related pages, which have diagrams that cannot be expressed in text), and you wouldn't just be saving a text file of the document contents either, but rather either a bunch of html/php files or files in a wiki markup language plus a parser. When all is said and done, I would be suprised if you could get the entire wiki into under 15-20MB, and would expect it to be more like 30-40. (Also, compressing 'raw text' (as you put it) rarely provides much reduction in size.)
  2. NAND memory is scarce enough that saving even 1 or 2 MB where possible is important.
  3. Things in the Firmware should only be those things that either almost everyone will use, or those that would be a pain to implement otherwise. This falls into neither category.
  4. The other points in my earlier post still stand.
 
Last edited by a moderator:
It could be useful if we maintain the wiki first.


For one: a lot of the information is horribly outdated.


Secondly: there is really poor structure. Many articles aren't well interlinked. Semantic MW could be a help here, but it would really take a lot of effort to really get things organized.


(I'm thinking: 5-10 very knowledgeable people working on it almost from the ground up)
 
(I'm thinking: 5-10 very knowledgeable people working on it almost from the ground up)
The problem is, people don't have that much time on their hands. Whenever I see something wrong on the wiki, I make a point of changing it, but to be entirely honest we're fighting a losing battle. I've been thinking recently that the community should have an 'update the wiki day' where where everyone pitches in, say, once every six months.


Either that, or find a way to clone Esn.
 
The problem is, people don't have that much time on their hands.
Exactly.

I've been thinking recently that the community should have an 'update the wiki day' where where everyone pitches in, say, once every six months.
That's sort of what I was thinking as well, but currently the wiki is so fragmented and outdated. It's really a general problem with wiki's.


I'm currently trying to maintain/update/improve a number of private wiki's and it's similar problems along the way.


Even if you have stuff organized from the ground up, at some point people come by and make some pages/info that's not organized and the story starts again.
 
Back
Top