Talk:Project Xanadu/Archive 1
From Wikipedia, the free encyclopedia
I think the text on the page for Project Xanadu needs to be edited to reflect the fact that MediaWiki nowadays actually use transclusion, and that this principle is thus in widespread use. Anyone else that agrees? Harvester 14:02, 5 Jan 2004 (UTC)
- Yes, I think it is a good idea to mention the transclusion/MediaWiki/Project Xanadu connection in the Project Xanadu article. Bevo 14:17, 5 Jan 2004 (UTC)
This is false. MediaWiki uses macro templating, not transclusion.--152.78.61.132 15:11, 9 March 2007 (UTC)
- It is termed transclusion, but it is not the same implementation, and the distinction is arguable. I would not consider it tranclusion in the xanadu sense, but it has the same functionality aside from the fact that it is essentially embedded. If one considers the html img tag to be transclusion, then one should consider MediaWiki's version to fit the definition as well, in my opinion. --John Ohno 12:26, 6 March 2008 (EST)
I haven't looked at anything Xanadu/Udanax related in about 2 years, but what makes it stick out in my mind since I started using Wikipedia is how similar Wikipedia is to the original vision; reverse linking, the version comparison feature looks identical (to my recollection)... -- Jim Regan 04:57 17 Jul 2003 (UTC)
We need some stuff on why Xanadu failed, and the Web succeeded.
Was it that:
- Xanadu was just too far ahead of its time?
- Xanadu was just too ambitious?
- the Web had the magic "worse is better" factor?
- Tim Berners-Lee "just did it", instead of talking about it?
- the Web built on existing SGML and FTP technologies, leveraging existing work -- Xanadu was "designed too soon" to be re-hacked on top of ad-hoc stuff like the Web.
-- The Anome 10:59, 5 Aug 2003 (UTC)
- I would personally consider it an anomoly of connectivity. Tim Berners-Lee had more connections to people who were supernodes (so to speak) in terms of mass adoption, and he oversimplified the design and implementation in order to dumb it down so that these supernodes would accept it when they had rejected the Project Xanadu version. Another element could be the fact that he was using it with the Internet (in the technical sense of the word -- as opposed to other widespan networks of the time, such as BBS and UUCP-based networks; it turned out that the TCP/IP stuff took off, and the web depended upon that, whereas Project Xanadu was (iirc) less specific at the time about the lower level protocols used). The "just doing it" factor may come in, but it relates directly to the complexity of the idea -- Ted Nelson had a pretty complex, nuanced idea, with its own terminology and such, and people generally didn't want to learn the nuances and thus were driven away. But note that this is my interpretation of the circumstances and events -- someone who was actually there, or who has better details, may be better equipped to elucidiate. Tim Berners-Lee actually discusses these points in Weaving the Web, and Ted Nelson discusses the same points elsewhere (I forget where -- somewhere in one of the documents on the old Xanadu site; I have caches in case the documents were taken down entirely).
-- John Ohno 12:35, 9 March 2008 (EST)
or that Xanadu seems horribly complicated? If I want to publish a document online, I just want to write it, not muck about with transclusion. In fact, transclusion sounds a bit like HTML frames: you need actual documents and a meta-document. There's a reason why frames died out! -- Tarquin 11:02, 5 Aug 2003 (UTC)
- The transclusion mechanism is supposed to be as automatic as the mediawiki-type "transclusion" when editing a wiki article. Transclusion is typically done either between versions of the same document, or between related documents, and it is a backend mechanism (not a frontend mechanism) which would ideally be as easy as a copy-paste (in fact, in one of his documents (maybe "Way Out of The Box"? Maybe something else) Ted Nelson states that Project Xanadu's concept of a copy/paste mechanism was actually a transclusion on the backend. The fact that Project Xanadu is formally against embedded markup would make the frame analogy innacurate, since the problem with frames comes from needing to create the metadocument and specify the documents to include, what pieces, etc in markup form (and also with the arguably naiive implementation of frames as compared to transclusions).
-- John Ohno 12:42, 9 March 2008 (EST)
-- LarryW 11:21, 5 Aug 2003 (UTC) I found these Wired articles interesting:
This is too self-referential to include in the article, but among the features which were available in Xanadu which can be seen in Wikipedia's software are:
- bi-directional links (What links here)
- Comparison of pages (the diff feature in the page history)
- Inclusion of files ("transclusion") ({{msg:}} feature)
-- Jim Regan 04:28, 14 May 2004 (UTC)
- Whoops. Mentioned that before! -- Jim Regan 04:32, 14 May 2004 (UTC)
NOTE: Deleted a completely erroneous paragraph claiming Xanadu was designed before computer networking and was intended to be centralised rather than distributed. This is simply not true, as the Xanadu design always featured distributed back-end storage with a back-end to back-end protocol. Andrew Pam, Chief Scientist, Project Xanadu.