Sunday, May 17, 2009

Bye Bye PBwiki, Hello Confluence

After having PBwiki host the wiki for our reference staff for the past four and a half years, I'm finally taking the plunge to move to whole thing (499 pages!) over to a locally installed version of Confluence. The college where I work got Confluence to use for various intranets needed around campus. Worried that some day PBwiki might just plain disappear, I decided to move the reference wiki over to something that is on our own servers and more under our control. It will take a chunk of the summer to laboriously recreate the wiki via copy and paste (wiki software from different vendors don't seem to make it easy to do import/exports). So much for that long week at the beach... 

6 Comments:

At 10:35 PM , Blogger Dorothea said...

Might it not be faster to send a webcrawler at the wiki and then search-and-replace the HTML into the new wiki's markup?

 
At 9:18 AM , Blogger Stephen Francoeur said...

I'm afraid that I am not quite web-savvy enough to understand your suggestion. In taking the content from PBwiki by the rather crude method of copy/paste, I get all the text but lose all the internal and external linking. I must admit to being ignorant about how a webcrawler can be used independent of a search engine. If I understand you correctly, I can use the webcrawler to capture all the content in its full HTML glory and then somehow take that file (or set of files) and do a find/replace with the file (or set of files) in Confluence? Will a webcrawler work in PBwiki if it is a password protected site?

 
At 1:46 PM , Blogger Connie Crosby said...

If you go into the edit mode for each page in PBWiki/PBWorks, and then click the button for source code, you can copy the HTML which will include all of your linkage if you want to hand copy it.

I'm not so worried that PBWiki will disappear, but that what we were doing for free is going to cost a lot. It's always a good thing to host your own, though.

 
At 2:14 PM , Blogger lalcorn said...

Stephen, is there any way to view the HTML source code and copy/paste that in as HTML, so you don't have to do so much work?

Nonetheless, yay you for making a reference wiki work. I can't get buy-in here.

 
At 2:51 PM , Blogger Stephen Francoeur said...

Connie and Louise: I noticed that when I copy the HTML from individual PBwiki pages and paste it into Confluence (either it the Rich Text Editor or in the Wiki Markup editor), the HTML doesn't get translated to the wiki markup for Confluence. I realize that I can then clean up each page and recreate the links, but I've got 500 pages to move, so if I can find a good way to do this as a batch rather than one page at a time, I'd be much happier. Looks like there may be some way that Confluence can convert HTML into its own markup, but I'll have to speak to the admin here on campus to have that plugin installed.

 
At 2:52 PM , Blogger Rikhei said...

Stephen, even if you can't do what Dorothea's suggesting, can you copy the source of the pages instead? That way you ought to at least be able to save the external linking.

 

Post a Comment

Links to this post:

Create a Link

<< Home