-
Notifications
You must be signed in to change notification settings - Fork 44
Description
I'm actually using the Python textile library, and filed a ticket on this there (textile/python-textile#16) but they (rightly) take the PHP version as the standard to follow. So I come before you with this humble feature request.
If you self-link a URL like
"$":https://github.com/textile/php-textile
the https:// scheme gets dropped from the linked string:
github.com/textile/php-textile
This seems less than ideal to me -- I think seeing the scheme really helps the reader "recognize" the string as a URL.
This gets to be more of an issue when you get into very long URLs (which is where the self-linking really comes in handy). Compare:
Data is available via web service. This is basically a block of text to bulk up the content and make the URL stand out a bit less. You can see an example of the data output by entering
service.example.com/ws/version/1/query?net=IU&sta=ANMO&loc=00&cha=LHZ
in your web browser.
vs
Data is available via web service. This is basically a block of text to bulk up the content and make the URL stand out a bit less. You can see an example of the data output by entering
http://service.example.com/ws/version/1/query?net=IU&sta=ANMO&loc=00&cha=LHZ
in your web browser.
When I read the first example, I feel like my eye initially "scans" the link as text, and passes the content on to my brain to read. But the content doesn't read as text, so my brain has to stop and go over it again more carefully.
Whereas the second one scans immediately as a URL, and my eye just skips immediately to the next block of text, and there's much less of an interruption in the reading process.
This is a fairly minor issue for an individual link, but as you might guess from the example one of our use cases is documentation of web services, in which case we have lots of (long) URLs mixed in with text, and having them visually blend together becomes a bigger problem.