4
$\begingroup$

Consider the three options:

  1. Source
  2. Source: NASA TM X-2525
  3. Source: NASA TM X-2525: Atlas-Centaur AC-17 Performance for Applications Technology Satellite ATS-D Mission

where in each case the linked url is less than self-explanatory https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720017275.pdf

I tend to use option #3 because it provides the most information to our customers the readers, but that's just me. However I've probably used [Source](url) before as well with Wikimedia urls and I'll have to curtail/undo that now.

I'd like to suggest that at least option #2 is much better for the answer's integrity and value to the site than option #1, based on the following reasoning:

Links rot in general, and NASA has at least in the past done major and minor revisions to it's websites and filing systems. I've experienced that myself, seeing swaths of links disappear or end in 404's, and over the last few years seen it referenced (lamented) more than once in other, more experienced users' comments.

Without at least the minimal NASA TM X-2525 reference, if the link breaks all you have is the following cryptic dead url:

and 19720017275 may not be sufficient to repair the answer.

$\endgroup$
11
  • 3
    $\begingroup$ anonymous close voter: this is meta and opinions about how the site is used are welcome here. $\endgroup$
    – uhoh
    Commented Aug 15, 2019 at 1:34
  • 2
    $\begingroup$ Probably the "leave open" flags will win. $\endgroup$
    – peterh
    Commented Aug 15, 2019 at 2:04
  • $\begingroup$ nasa and 19720017275 ARE sufficient to get the document from the Internet Archive $\endgroup$
    – user20636
    Commented Aug 15, 2019 at 8:36
  • $\begingroup$ While not mine, I suspect the VTC is because it's a matter of personal preference. $\endgroup$
    – user20636
    Commented Aug 15, 2019 at 8:42
  • $\begingroup$ @JCRM this is just an example, the question is about best practices $\endgroup$
    – uhoh
    Commented Aug 15, 2019 at 8:47
  • 1
    $\begingroup$ to some it could seem like an attempt to turn your preference into a defacto site rule. $\endgroup$
    – user20636
    Commented Aug 15, 2019 at 8:53
  • $\begingroup$ There are no such thing as "site rules", nobody reasonable would think that even for a second. To some suggesting I'm attempting to turn my preference into a de facto site rule could in turn seem like an attempt to raise a false flag. $\endgroup$
    – uhoh
    Commented Aug 15, 2019 at 9:00
  • 2
    $\begingroup$ @JCRM In that case they should downvote the question, but it is clearly on topic for meta. Close is out of line. $\endgroup$
    – called2voyage Mod
    Commented Aug 15, 2019 at 13:35
  • 1
    $\begingroup$ This is a good point, if anyone wants to check their own answers, here is a SEDE query: data.stackexchange.com/space/query/1336294 apparently, I better update three of my answers. $\endgroup$
    – Polygnome
    Commented Nov 24, 2020 at 13:42
  • $\begingroup$ @Polygnome to improve visibility I think that could be posted as a real answer. It helps people find their own links then they can decide if the URL contains enough information to reconstruct or search for a new link or not. $\endgroup$
    – uhoh
    Commented Nov 24, 2020 at 13:49
  • 1
    $\begingroup$ @uhoh I'm currently still tinkering a bit with the query because it doesn't work too well, yet, as I've since discovered. Once I'm done I'll consider it. $\endgroup$
    – Polygnome
    Commented Nov 24, 2020 at 14:11

2 Answers 2

4
$\begingroup$

Yes, they are "sub optimal" because a URL + title gives more information than just the URL. You have given the reason yourself *.

On some sites the 'optimal' link would even give more information. I'm specifically thinking about Skeptics where notability and dates are important 'meta-information'. There I sometimes edit links to include medium/source and date, like this:

Here’s how Apollo 11 forever changed how we watch TV (Washington Post, July 20, 1969)

Now, how much time you want to spend on improving link (=site) quality is up to you, of course ;-)

* There is one thing that may help to fix broken links: The Wayback machine at archive.org

$\endgroup$
5
  • $\begingroup$ does the Wayback machine often fix broken links, or just help prevent them if you think of it ahead of time? $\endgroup$
    – uhoh
    Commented Aug 15, 2019 at 8:18
  • 1
    $\begingroup$ No it does not fix anything. You can try to go to earlier versions of the page that is gone, and then use their link instead of the original. $\endgroup$
    – user10509
    Commented Aug 15, 2019 at 8:24
  • 2
    $\begingroup$ If you think of it ahead, you can suggest the Internet Archive has a copy of your link. $\endgroup$
    – user20636
    Commented Aug 15, 2019 at 8:32
  • 3
    $\begingroup$ I always capture title and organization/site or author name for my explicitly notated Source section, but I sometimes have links in the body of my text that I don't take this care with. Ideally those should be additional info, but I know in many cases I have linked sources that way. I have switched links to Internet Archive copies before, but I wish I were better about remembering to make sure links were already in the Wayback Machine first, lol. $\endgroup$
    – called2voyage Mod
    Commented Aug 15, 2019 at 13:39
  • 1
    $\begingroup$ @called2voyage That last thing is definitely something that 99.99% of SE users have not even thought about $\endgroup$
    – user10509
    Commented Aug 15, 2019 at 14:24
1
$\begingroup$

This is a good time to remember why it is important to make sure all the relevant information from your link is in your answer. If links ever go down and the original source material cannot be found, it is better to have second-hand information than nothing. As long as Stack Exchange stays up, the second-hand information will persist for people to use.

$\endgroup$
0

You must log in to answer this question.

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