Will TW5 be Google friendly?
107 views
Skip to first unread message
kev
Apr 26, 2012, 6:08:06 AM4/26/12
to TiddlyWiki
I have posted now and again about the difficulty Google has indexing a
Tiddlywiki. Will Google be able to index the new version? Or is that
asking too much?
If Google could index all the content in my TWs I probably wouldn't
have created my Mediawiki site. Why bother when TW is so good.
Also a proper newsreader plugin would be great. :)
Tiddlywiki. Will Google be able to index the new version? Or is that
asking too much?
If Google could index all the content in my TWs I probably wouldn't
have created my Mediawiki site. Why bother when TW is so good.
Also a proper newsreader plugin would be great. :)
Tobias Beer
Apr 27, 2012, 1:15:45 PM4/27/12
to tiddl...@googlegroups.com
Hi kev,
+10
Would you mind sharing your idea of a 'proper newsreader plugin'?
Cheers, Tobias.
Jeremy Ruston
Apr 27, 2012, 2:56:36 PM4/27/12
to tiddl...@googlegroups.com
It's a good question. The
short answer is "yes".
The medium answer is that exposing your content to Google is a fairly intricate business whatever tools you use, and TiddlyWiki5 aims to give you a high degree of control over what Google sees.
For the longer answer, I should first explain the background to the problem. TiddlyWiki, by definition, stores content in wiki text format rather than conventional HTML. It works by using JavaScript to convert the wikitext to HTML on the fly when it needs to be displayed.
The fundamental problem is that Google (and other search engines) don't understand wikitext. Classic TiddlyWiki tries to ameliorate the problem by storing the wikitext as if it was HTML, so that Google at least sees it. In many cases this means that Google will be able to index key words, phrases and URLs within the content. (For example, if I search Google for the URL http://tiddlywiki.com/firstversion.html, the second match is an extract of the raw wikitext from tiddlywiki.com).
There's no simple solution to the problem. TiddlyWiki can't use conventional HTML to store content because it doesn't have the ability to represent wikitext structures. (The fundamental issue is that two way conversion from wiki text to HTML and back again is only practical in limited circumstances).
Serverside solutions like TiddlyWeb and TiddlySpace avoid the problem by using the TiddlyWiki engine to give Google HTML versions of every tiddler.
Similarly, in TiddlyWiki5 the problem goes away when running as a server. When running as a single file in the browser, the plan is twofold:
* When publishing, to provide the option to save a custom representation of some or all of the content as HTML, embedded within the HTML file such that Google will see it
* To allow ordinary wikitext content to be moved into special script blocks, hiding them from Google
(This solution is actually very old in TiddlyWiki terms; there are various splash screen plugins that work this way).
The custom HTML representation could be configured to be just a list of the title of each tiddler, or the full text of each tiddler. The tradeoff is that the content is effectively duplicated within the file, which may limit it's applicability for large data sets -- which is of course where the server side solutions come in).
Anyhow, none of this functionality is implemented yet, but will be soon. There will still need to be a certain amount of testing to ensure that we're presenting things to Google as optimally as we can.
Best wishes
Jeremy
The medium answer is that exposing your content to Google is a fairly intricate business whatever tools you use, and TiddlyWiki5 aims to give you a high degree of control over what Google sees.
For the longer answer, I should first explain the background to the problem. TiddlyWiki, by definition, stores content in wiki text format rather than conventional HTML. It works by using JavaScript to convert the wikitext to HTML on the fly when it needs to be displayed.
The fundamental problem is that Google (and other search engines) don't understand wikitext. Classic TiddlyWiki tries to ameliorate the problem by storing the wikitext as if it was HTML, so that Google at least sees it. In many cases this means that Google will be able to index key words, phrases and URLs within the content. (For example, if I search Google for the URL http://tiddlywiki.com/firstversion.html, the second match is an extract of the raw wikitext from tiddlywiki.com).
There's no simple solution to the problem. TiddlyWiki can't use conventional HTML to store content because it doesn't have the ability to represent wikitext structures. (The fundamental issue is that two way conversion from wiki text to HTML and back again is only practical in limited circumstances).
Serverside solutions like TiddlyWeb and TiddlySpace avoid the problem by using the TiddlyWiki engine to give Google HTML versions of every tiddler.
Similarly, in TiddlyWiki5 the problem goes away when running as a server. When running as a single file in the browser, the plan is twofold:
* When publishing, to provide the option to save a custom representation of some or all of the content as HTML, embedded within the HTML file such that Google will see it
* To allow ordinary wikitext content to be moved into special script blocks, hiding them from Google
(This solution is actually very old in TiddlyWiki terms; there are various splash screen plugins that work this way).
The custom HTML representation could be configured to be just a list of the title of each tiddler, or the full text of each tiddler. The tradeoff is that the content is effectively duplicated within the file, which may limit it's applicability for large data sets -- which is of course where the server side solutions come in).
Anyhow, none of this functionality is implemented yet, but will be soon. There will still need to be a certain amount of testing to ensure that we're presenting things to Google as optimally as we can.
Best wishes
Jeremy
--
You received this message because you are subscribed to the Google Groups "TiddlyWiki" group.
To view this discussion on the web visit https://groups.google.com/d/msg/tiddlywiki/-/lYOE_jHE3nsJ.
To post to this group, send email to tiddl...@googlegroups.com.
To unsubscribe from this group, send email to tiddlywiki+...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/tiddlywiki?hl=en.
kev
Apr 28, 2012, 3:51:06 PM4/28/12
to tiddl...@googlegroups.com
Thanks for such a complete reply. From a laymans common sense point of view, It seems as much a shortcoming of the Google search engine as much as a TW "problem". I've got web space so I look forward to using the server function. My 5 year old TWs never have been properly indexed!
As for the newsreader the old rssReader worked perfectly but the author stopped developing it and now it doesn't work with the most recent Firefoxes. It was very useful to read my newslists from within TW using a purpose built newsreader. I now use an embedded version but that will depend on the source provider always being available.
As for the newsreader the old rssReader worked perfectly but the author stopped developing it and now it doesn't work with the most recent Firefoxes. It was very useful to read my newslists from within TW using a purpose built newsreader. I now use an embedded version but that will depend on the source provider always being available.
To unsubscribe from this group, send email to tiddlywiki+unsubscribe@googlegroups.com.
Jeremy Ruston
Apr 29, 2012, 10:04:47 AM4/29/12
to tiddl...@googlegroups.com
As for the newsreader the old rssReader worked perfectly but the author stopped developing it and now it doesn't work with the most recent Firefoxes. It was very useful to read my newslists from within TW using a purpose built newsreader. I now use an embedded version but that will depend on the source provider always being available.
Yes, I like the idea of having ones one customised RSS Reader. I'd hope that we'd end up with an RSS Reader plugin for use in serverside TW5.
Cheers
Jeremy
To view this discussion on the web visit https://groups.google.com/d/msg/tiddlywiki/-/rFVzsG4AZDEJ.
To unsubscribe from this group, send email to tiddlywiki+...@googlegroups.com.
Jeremy Ruston
mailto:jeremy...@gmail.com
Reply all
Reply to author
Forward
0 new messages