6

I've been reviewing a number of edits which consist only of changing plain link URLs to URLs hidden with descriptive text. For a fictional example:-

You can read more about this here: http://www.interestingbikeblog.com/wheel-noise

edited to:-

You can read more about this on the Interesting Bike Blog: Wheel Noise

It seems to me that often it's helpful to see the URL without having to hover over it, especially when it's as descriptive as my fictional example above.

Sometimes the URL is a shortlink which isn't descriptive, but even so the edit to add text doesn't seem to add much value and could be considered trivial.

What is the recommendation on reviewing these edits?

4

I have just developed a strong opinion on this :) The edit to this answer that I fixed was not an improvement, but demanded effort from us. That makes it negative. This edit similarly fixed only one of two possible problems, but somehow that too was approved. I suppose an arguable half fix is better than nothing?

Edits as above that leave link obfuscation sites in the chain are bad. The whole value of the edit is, as Batman says, that we can have as much link text as we want. That makes the URL "shortener" merely an unnecessary point of failure. It suggests that the person making the edit is not trying to improve the site, they just want points.

In the case of link-only answers the offence is doubled. Not only is the link mangled, it's not even excerpted per the guidelines. So the "editor" has not even RTFM.

I thus suggest rejecting those edits where possible, specifically, "reject and improve".

  • 1
    URL shorteners should be avoided for all kinds of reasons (unnecessary point of failure, makes it hard to judge whether the link is safe or appropriate to follow, etc.) so an edit that removes a URL shortener is surely a good thing. Edits that introduce URL shorteners should be rejected. – David Richerby Oct 12 '16 at 9:13
  • But, beyond that, I really don't understand what you're saying. An edit that "only" improves one aspect of a post should be rejected? I agree that somebody who submits multiple small edits to the same post instead of one big edit is probably angling for rep but we can deal with that by flagging and asking the mods to have a word. If an edit makes a genuine non-trivial improvement, "it doesn't make the post perfect" is a nonsensical reason for rejecting it. Rejecting it is just as much work as accepting it and leaves the post in a worse state than accepting it. – David Richerby Oct 12 '16 at 9:16
  • You seem to believe that anybody who produces an edit that is not perfect is "trying to game the system." I don't understand that claim. – David Richerby Oct 13 '16 at 7:02
  • @DavidRicherby I think you should write an answer that argues for your position, rather than simply arguing against mine. – Móż Oct 13 '16 at 22:31
  • I prefer raw links because I like seeing the actual url before clicking on it. – BSO rider Oct 15 '16 at 21:44
4

I think its a trivial edit in some cases, but in some cases it can improve readability (and in those cases it should be approved).

For example, if I have the following link inline http://www.trekbikes.com/us/en_US/bikes/city-bikes/urban-commuter-bikes/crossrip/crossrip-2/p/1376000-2017 (or something longer) it might impede readability and shortening it with text may be useful. On the other hand, if its something short like http://www.trekbikes.com, probably not.

I'd go on a case by case basis.

1

Hypertext was originally intended to hide the addressing details. So from that point of view putting a text label on the URL is a good thing.

On the other hand, if I see the link and realise its to sheldonbrown.com O might have a different desire to click it compared to foamingmouthhippy.org

If its the only edit, then the question pops up to the top of the active list, even though its not really active.

I'm torn.

  • I don't think bumping questions to the top of the list is a huge issue on this site, as long as we're only talking about a couple of posts here and there. We're sufficiently low-volume that moving a couple of questions to the top isn't going to push truly active questions off the bottom. Right now, there are only 19 questions that have been modified in the last 24 hours, and another twelve from the last 48 hours. I have it set to display 50 questions and the oldest of those haven't been touched in a week. – David Richerby Oct 12 '16 at 9:21
1

One thing that hasn't been mentioned is search indexing.

Search engines are better able to index our site when we use descriptive link text. Phrases like "Sheldon brown talks about chain tension" are more beneficial than "read about it here: http://www.sheldonbrown.com/NEW-chain-adjust.html." When someone Googles "chain tension," the first will bring up the thread on our site, the second won't. The former also helps Sheldon (or whoever) because search engines take that descriptive text into account when ranking the linked site.

In order to drive more traffic to our site, I would recommend using descriptive links whenever possible.

URL shorteners are a different issue. While we don't have a hard and fast rule about them, my personal opinion is that they don't belong anywhere on the internet except Twitter*. URL shorteners make it so you have no idea what you're clicking on until after you have clicked. That short URL could be an innocent blog posting, a virus download, or child pornography. You just can't know until after the fact.

*The only reasons I find them acceptable on Twitter are 1) the 140 character limit and, more importantly, 2) I don't use Twitter.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .