Blogger users: if you don't hyperlink an URL it isn't linked on web browsers but *is* on mobile version

by @JoBrodie, brodiesnotes.blogspot.com

Occasionally I refer to a web address without linking it in the (now mistaken) belief that this doesn't give it any 'Google juice'. This is because by not hyperlinking (making it clickable) visitors can't click on it to go to that site, so there's no trail of clicks from my site to the site I'm mentioning and nothing for Google to index.

I do this for iffy websites that I don't want to link to but where I want to provide the information for people to be able to visit the site if they wish (they can copy and paste the address into a new browser tab).

However it doesn't work. When you're viewing this site on a mobile phone any unlinked address automatically appears as a clickable thing.

On a browser this looks like plain text
http://www.google.co.uk - I have not hyperlinked this one, but I have hyperlinked all other links in this post.

but on a mobile version it would like
http://www.google.co.uk

(if you're viewing this post on a mobile you'll see both hyperlinked, check on a desktop browser to see what I mean).

Solutions?
1. Mangle the URL in an obvious way so that the reader can see what to do to in order to visit the page, eg http://www.goDELETEoogle.co.uk - click all you like, you'll go to the wrong place until you delete the word DELETE from the URL.
2. Use Do Not Link* http://www.donotlink.com/ to create a referring address which acts as a sort of protective wrapper so that Google, in indexing links on your site, indexes the referring link and not the real one.

*"Who uses donotlink?
Skeptics, bloggers, journalists and friends on social media use donotlink to link to scams, pseudoscience, misinformation, alternative medicine, conspiracy theories, racist / sexist blog posts, etc. without improving the search engine position of the site they are discussing."

I should add that using Do Not Link doesn't actively harm the search engine position of the site, which is something that has been mis-claimed by Lynne McTaggart (editor of What Doctors Don't Tell You) in wondering if skeptics are trying to damage the magazine's Google reputation by using this style of linking. This is not what's actually happening, by linking in this way we're not benefitting the site, but it's not otherwise harmful.

What is all this crap Google adds to my search URLs?

I am a habitual URL hacker* - tinkering with the bits in a web address to try and find a page that I want. For example if I'm on website.com/section/interestingpage.html I might prune off the bit at the end and see if there's anything at website.com/section - it's often brought up loads of great stuff (not always, sometimes you don't get permission to view intermediate bits).

Whenever I run a search on Google and share the results with someone else I've noticed that I always paste the results URL into notepad and clip off all the stuff at the end that appears to be completely irrelevant to the search. I've only just started noticing that I do this although I've been doing it for as long as I can remember. I do it for everything though - if I'm sharing a link to an article on a news site and it has a bit at the end that mentions the referrer link (something like utm=twitter I think) then I prune that out too. I'm not exactly sure why, unless it turns out that I'm a purist and only want to share the minimum of alphanumerics needed to direct someone to the right page on a site.

The URL clipping itself is the work of seconds as I always have a notepad document open for that sort of thing and use keyboard shortcuts Ctrl+C and Ctrl+V to copy and paste and Alt+Tab to ping into the notepad window.

I've just searched on my work website for info about 'diabetes networks' using this exact search:
site:www.diabetes.org.uk "diabetes networks" - the site: bit means it just searches within our website and the bit in "" marks means it searches that exact phrase.

Wanting to send the results to someone else I copy the URL from the address bar and, as always, note with surprise the random string added to the end - what is all this guff that Google adds to my search string?

https://www.google.co.uk/search?hl=en&q=site%3Awww.diabetes.org.uk+%22diabetes+networks%22&oq=site%3Awww.diabetes.org.uk+%22diabetes+networks%22&aq=f&aqi=&aql=&gs_sm=3&gs_upl=337562l337562l0l337807l1l1l0l0l0l0l87l87l1l1l0&gs_l=serp.3...337562l337562l0l337807l1l1l0l0l0l0l87l87l1l1l0.frgbld.


The URL above contains a duplicate of the search string (in dark red) and a whole load of apparently non-contributing filler (in green).

Deleting all of that pares it back to this much smaller web address: https://www.google.co.uk/search?hl=en&q=site%3Awww.diabetes.org.uk+%22diabetes+networks%22 - which works fine. The %22 is the way web addresses render ", so you need to keep those (%3A = : - the colon between site and www).

The aqi / aql / gs_sm stuff is occasionally useful - if you click on the image search for example, you're given in return an address that looks a bit like this and removing these bits of text from the URL just resets it to the regular web search (meaning if you send the amended link to someone else they'll just see the regular results, not the image results).

But what do all those numbers and letters in green mean?

*solely for benign purposes!

Rarely a fan of embedded links within low-information phrases

I like information and I like organising it and, to me, a web address (URL) contains information.

Even if an URL is embedded like this (Nicola Conte's "Jet Sounds" on YouTube) you can still see what the address is by hovering over it (unless you've switched off that particular status bar) - this just lets you know where you're clicking to, can be useful if you suspect a spam address etc.

Sometimes URLs don't contain much useful information though and I've just found a small example of some unhelpful addresses in a piece of text I wrote, but which is fortunately saved by me having written the name of the file into the text.

I have a secret stash of pointers to diabetes statistical information but... because the Office for National Statistics have recently tweaked their website this means that virtually none of their links work or redirect to ones that do, so the information below goes nowhere. I'm patting myself on the head for at least having included the document titles so I'm fairly confident I'll be able to find the files' new location.

This is exactly the same as supermarkets moving the store contents around (to get you to take a different path and buy more stuff you've not seen before) but at least they usually produce wee leaflets to guide you to where everything is.

"Data for “Deaths ONS” column (Office for National Statistics) for 1999-2005 taken from spreadsheets in Mortality Statistics: Cause (Series DH2), 2006-2009 data is from Mortality statistics: Deaths registered in England and Wales (Series DR) and 2010 from Death registrations in England and Wales, selected data tables (specifically “Death registrations by sex and single year of age, 1961 to 2010, England and Wales and UK”)."

I'll be spending a small bit of time finding the new location for these and will post them up as I come across them. I'm not too worried as even if I don't find them myself the @statisticsONS Twitter people are lovely and helpful and have reunited me with several much-loved stats files already.

Edit
I went to ONS pages, clicked on the Data tab and then searched for 'Mortality statistics' - looks useful.

Another good route in is the Browse by theme tab, then on the right hand side of the page look for the 'Causes of death' link within the Health and Social Care / Health of the Nation bit.

Yes I have been a bit careful about embedding these links :)

Mortality Statistics - deaths registered in England and Wales (Series DR) 2009
http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2009/index.html - I'm also leaving the raw URL here as the nice thing is that you can copy and paste this and replace 2009 with 2008. If only everything was this simple!

In all cases the relevant one, for diabetes deaths, is Table 5.4

http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2010/index.html

http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2009/index.html - once there choose Data in this release and then look for 5.4
- eg Table 5.4 Deaths: underlying cause, sex and age-group, 2009: Chapter IV Endocrine, nutritional and metabolic diseases

http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2008/index.html

http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2007/index.html

http://www.ons.gov.uk/ons/rel/vsob1/mortality-statistics--deaths-registered-in-england-and-wales--series-dr-/2006/index.html

That's as far as it goes...

The printed word and http://

When I'm looking at a page of printed text I find it a bit hard to spot the web addresses embedded within it unless they are in colour or underlined. Web addresses written with www. at the beginning, instead of http://, just seem unfinished to me, and perhaps less easy to spot. An http:// or two is helpful to guide the eye, being the textual equivalent of a highlighter pen. A sensible alternative would be to put the web address at the end of the text and reference it in-text, and if it's a long address it's helpful to provide a shorter alternative to save whoever's typing it in from the printed page.

Some accessibility guidelines recommend against underlining text (presumably because the line interferes with the shape of the words but since we've managed to invent strikethrough and double strikethrough surely we can invent an underline that's lower down...) or italicising words, meaning that the links aren't made particularly salient. Sometimes links are presented in bold, which is a big help, but it depends on the house style.

This is when I skim read in hope of finding http:// as this 'word' looks different from all the other text on the page. It's probably the :// that's doing it but the h and the p form a nice shape too. Probably we've all got used to the idea that typing http:// into a browser address bar is generally redundant as the default setting is for web pages (although you have to be careful if you want to use the https:// format for secure pages).

In electronic documents internet addresses are usually more obvious. Pressing the space bar after typing one in my email programme (Outlook) or Word automatically turns the preceding address into a link or email address (of course you can set it so that it doesn't do this). In some email programmes the mailto: command isn't needed and the presence of an @ is enough to create a clickable email address, similarly the http:// isn't necessary and www.blah will resolve to an active link.

Plenty of web addresses don't even need the www, for example bit.ly and twitter.com. Having said that I notice that the http:// seems still to be required in tweets in order to render an address as a blue-text active link (similarly # in front of a word renders it as a link to the Twitter search results for that word).

Presumably someone can 'set' Twitter so that bit.ly/blah will go to http://bit.ly/blah in the way that someone must have done something to email a few years ago so that mailto: in front of an address would 'activate' it. But at least in these cases it's obvious what the link is, whatever it looks like.

But for printed pages, bring back the http://

With thanks to @the_beacon for the point about https and to @jjsanderson, @mokuska and @rpg7twit for disagreeing with me :)