Jump to content

Make location URLs more future-proof


jlm
 Share

Recommended Posts

My notes (outside of Accordance) now have thousands of accord:// links, and I've begun to worry about their future usefulness. My concern is that they're linking to Tools with a paragraph number, counting from the beginning of the module. If a correction splits or joins paragraphs, or adds some content that was initially left out, then paragraph numbers will change, by a little or a lot, and a link to paragraph 8319, for example, won't be the same paragraph that had that number when I copied a link to it (or will it be? Are these numbers already stable?).

 

An even greater change takes place when volumes are added to a commentary. In this case the module name changes, and I don't suppose links to the old module will open the new one to the corresponding location. I could search and replace to change the module name in my links, but the paragraph numbers will only correspond for paragraphs prior to the first new volume added to the set.

 

So I suggest that location URLs be offsets from something closer to the place linked than the beginning of the module. This could at least be the beginning of a volume, e.g., instead of accord://read/WBC-NT-25#18029, accord://read/WBC-NT-25#35B.1712 (meaning volume 35B, paragraph 1712). Better still, count from something even closer, like a Reference location, the name of a dictionary article, or a page number. Volume, page number, and paragraph offset could be a good way to make a stable link (accord://read/WBC-NT#35B.553.4). With the volume number, we no longer need to add the volume count to the module name, and the link could keep working when volumes are added. WBC-NT-25 just has to know it provides the resource "WBC-NT", as a future WBC-NT-27 might.

 

It seems that Logos has noticed the issue, because they've introduced a new URL format, although I think including snippets of text to help locate the passage is going a bit overboard.

 

I think it would also be a good idea to adopt something more like a URL query syntax, where parameters are separated with ampersands and standard URL encoding is used to protect special characters. The first benefit of URL encoding would be that it would eliminate the ambiguity between underscore and space, making it possible to link to books that have underscores in their names.

accord://read/CLASSICS-E#5Ar_Nic._1137:1 (error due to ambiguity) -> accord://read/CLASSICS-E#5Ar_Nic.+1137:1 (unambiguous)

 

It would also make it possible to specify additional parameters needed to make a search work:

accord://search/WBC-NT-25;Page_Numbers?35B_553 (copied URL doesn't indicate flex is needed) -> accord://search/WBC-NT-25?35B+553&field=Page_Numbers&flex=1

 

And perhaps we could specify additional modules to show in parallel, like accord://read/NABRE?Isa.+66:10-14&parallel1=HMT-W4&parallel2=Anchor_Isa-Mal

Platforms that only allow one parallel resource (like mobile devices) would ignore additional parallels.

 

More thought is needed on how to make URLs as future-proof as possible.

 

  • Like 3
Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
 Share

×
×
  • Create New...