I agree with EarlyOut… I’ve grown accustomed to Ctrl+Clicking Google results, and get annoyed in the rare case when I forget to. _blank links wouldn’t change Ctrl+Click behavior, so would improve the Google experience for me personally. I know Early and I are not alone in this.
No one’s said “Nielsen is always right so you must be wrong.” No one is always right. And many of the valid arguments against “target=_blank” iin the above posts do not cite Nielsen. You’ve done what so many people do: take a single thread of a debate, criticize it, and say that because you’ve criticized that single thread, the entire schema of the debate is now in question.
Having said that, I appreciate your efforts to keep the discourse civil in this thread, and I have no issues with the way you or anyone else have stated their cases. It’s a healthy disagreement we’ve got going on here. Have at it, gentlebeings.
I hear what you’re saying Max but I haven’t seen a single valid “against” for the case of _blank links in UGC that takes my pov into account. All arguments against simply state that it’s wrong period, essentially quoting the Nielsen stance (whether intentional/explicit or not).
Sorry, but for only 2% more users I rather have valid mark-up rather than a new window opening. I mean… perhaps if that number were significantly higher I would reason with you but for only 2% not even worth the trouble. That means that design wise using JavaScript to enhance the experience will function for 98% of users, which would b more than fine with me. Opening a new window/tab automatically is in essence a design/usability enhancement, expected or not. The only time when its not so much a design enhancement is linking to things outside the domain for vendors, partners, etc. In that case the marketing people throw fits it seems but otherwise, opening a new window/tab is a usability enhancement. A enhancement that isn’t really worth invalid mark-up for a mere 2%. Than again I tend to always at least make sure primary pages of a site/application validate.
I would be SHOCKED to find that 2% give a flying fig about a new window opening or not… If they are too stupid to know about middle-click (I see ctrl-click mentioned, but no mention of middle? Thread full of Mac users or something?) I don’t think opening a window is really going to make a difference to them or not. I mean if they’re too stupid to realize a mouse has more than one button, shoving a non-standard behavior down their throat isn’t going to help with that.
God forbid clicking on a link to take you away from the current page takes you away from the current page!?! That’s what links do… having them open a new window or not willy-nilly is nonsensical, inconsistent, and annoying.
ESPECIALLY with #DDD IE users who end up with forty windows open and then kvetch about having to close all of them – assuming they ever notice. More than once I’ve gone into offices where people were complaining about the machine slowing down where they’ve had triple digits of IE windows open… “group by program” on the taskbar for the win – NOT… ALL because some {expletive omitted} decided to abuse TARGET for something it was NEVER actually meant to do. ( _blank is actually an invalid target name, it cascades through to a new window as an ERROR)
Though I am a little surprised to see people so attached to something so COMPLETELY POINTLESS. Especially when it confuses users by breaking forward/back navigation.
But of course, ignore the published standards, ignore the accessibility recommendations, ignore the recommendations of accessibility experts and just keep vomiting up broken half-assed code any old way. Just part of my increasing disgust with the sleazeball web development industry as a whole… Where 90%+ of the people making websites out there need to do the world a favor, back the {expletive omitted} away from the keyboard, and take up something harmless like macramé weaving.
All and all the average user (my mom) does not even notice if a link opens in a new window or not. So basically it boils down to how the maker of the site would prefer it to perform. Most everyone else could care less much less even notice.
Ah yes, but you’re in the elite. A lot of sheeple who search on Google think that the first link is the only link and everything else is just pretty decoration. And Google itself wants the first link to be the only one you need to look at too… (only slightly tongue-in-cheek)
Again, I find it less of an irritation to middle-click to open a link (which is no more or less effort than left-clicking) as and when I want to, rather than end up with surplus tabs open when I don’t.
I usually middle-click, as well, but I’ve got a Logitech mouse that has the annoying habit of not being able to decide whether a middle-click is “open in new tab” or “turn into a sticky scroller.”
Stevie, last I checked top 10 results had a click degradation rate around 30-40% from result to result. Not far below what you’d expect.
You left out the most obvious alternative. You can add your own userscript into your browser to tell the browser to open all links on Google pages in a new tab. Doing that will get the exact behaviour you want without interfering with other visitors who want to open their Google links in the same tab. (At least this option is available to you if you are using Internet Explorer, Firefox, or Opera).
The person who owns the browser is the one who should be making the decision as to what links should open where. Both Opera and Firefox (with the right extension) allow you to configure where you want certain types of link to open and userscripts can allow you to set behaviour on a site by site or even page by page basis. The only thing that can interfere with your doing this are those IDIOT web sites that insist on taking away one of the options by specifying target attributes (although you can set your userscript to strip those out but if they want to be that unfriendly toward their visitors then there are probably loads of other usability issues with their site as well - so a better alternative is to just strip out all the contentwhen the first target attribute is found so that you know not to waste time on their site).
No one is going to die if a page doesn’t validate. You might die, but I know the majority of people won’t. That isn’t to say I don’t support many of the things you have said, but really… we aren’t surgeons here and thinking so is a little arrogant. I mean… especially when it comes to front-end development. There isn’t much physical harm that can come from doing it the wrong way unless perhaps your the one who needs to maintain it. Now, depending on the project back-end programming (ie. cc processing) can be harmful, but not HTML. So a site is inaccessible or broken… whooptie f**kin doo!
While I agree with you the majority of people are pretty f**kin stupid when it comes to the web. In many cases forcing decisions for a user is a the right way to go. The less decisions a user has to make the better. Most users barely know how to open up “google”… Furthermore, they ain’t running Opera or Firefox but IE7 or at best (worst?) IE8. If they are technically capable than changes are they know how to customize their experience anyway.
I agree for the most part but sometimes its easier just to do what people want than fighting over something that is really so insignificant. Until someone complains though I generally do not open things in a new window unless its directly requested.
Now you’ve done it. When you’re dealing with the Taliban, heresy is a capital offense.
Generally all those issues are not directly the fault of the programmer but the consequences of using CMS that attempts to provide the most flexibility for styling items without modifying any server side code.
Nonsensical headings are normally the consequence of a CMS that attempts to make regions reusable – never knowing the context that they will be used in. One reason I am happy to see the authors making headings relative to certain elements, rather than the entire document. The only way a CMS that provides regions, block, etc to have meaningful heading levels would be the user would need to understand and define it some way themselves – not going to happen for most users.
THINK about what you just said… SERIOUSLY.
If nobody wants to use it because it’s inaccessible and/or broken, what’s the point of even making a site in the first place?
I just laugh at the idiocy that goes with it – that somehow fat bloated poorly written code that ignores all the rules and conventions is somehow easier to write and maintain… ON WHAT PLANET?!?
As to the CMS part, I’d buy that for the presentational class hooks, MAYBE… Still manure coding practice… But explain this type of idiocy:
<li class=“menu-243”><a href=“/services” title=“Services”>Services</a></li>
There’s no excuse for that other than “dooeey, I don’t understand what a title attribute does”
Or worse, wordpress’ idiotic “let’s slap the same class on everything” inside a perfectly good already classed parent.
I’m not saying there is a point. I’m saying in every industry there are those type of people, who don’t care about quality, slapping sh*t together and get the clients write of. What I am saying is at least there isn’t much actual harm that can come from it. That doesn’t justify it by any means… I’m just saying. So I’ve learned so long as I’m not the one maintaining it I could care less but when responsible for maintaining or creation your damn right I fight for use of proper practices and expect the same out of the people I work with. Which can be tough times working with Drupal and all the wonderful modules available, but that is another story entirely.
Don’t get me started on WP…
There are some CMS and I believe WP is one that provides an input for the title attribute. So again in that case its not really the developers fault, more like the users fault. Than again how can you expect the average user to understand the actual purpose from a technical stand-point. Yeah, perhaps it would be nice for systems to check if the supplied title is the same as the text within the anchor and provide a warning but from programmatic aspect its really not an error.
deathshadow60 – you would love Drupal views or better yet panels. Powerful, powerful things from a programmatic and user stand-point but my god is the mark-up terrible.
Let’s talk specifics before you fling around insults any more, buddy… SPECIFICALLY, let’s look at THIS failed validation:
[Invalid] Markup Validation of https://www.marijuanadoctors.com/ - W3C Markup Validator
It has exactly 4 errors, and I can’t correct them without modifying addthis and the underlying development framework. But I don’t give a flying fark, because none of these validation failures mean A THING to any browser I’ve ever seen. So where you might spend hours on end ensuring you get a pretty green box from the W3C, I’ll be on to the next project. I guess that makes me an idiot in your eyes… the only thing I feel like an idiot for is wasting my time explaining this to someone who obviously doesn’t get the “business” end of this business.
Reminds me of that classic definition:
consultant
- noun
- A person who knows 100 different ways of making love, but doesn’t know any actual women.
Off Topic:
Personally I’d be worried about the CAPTCHA it seems not to be needed for an account to be created I don’t know if that was intentional or not? But it allowed me to register with it being blank.
Ok lets keep this civil and back on topic which is about :
"[B] target = “_blank”.
Discuss issues as strongly as you like but just remember don’t make it personal or abusive.
[/B]
Funny since at a cursory glance I’m counting at least a dozen more – meaning the validator threw up it’s hands and walked away; a common occurrence when it hits two or more completely fictional attributes.
Like this malarkey:
<a class=“addthis_button_facebook_like” fb:like:layout=“button_count”></a>
Though this is consistent with that I’m seeing four different layouts in five different browsers, only two of which look even remotely functional. As such:
When it doesn’t even work, sucks down bandwidth for nothing costing it more to host, has zero accessibility, etc, etc, etc…
Enjoy your… what, 95% bounce rate? (ballpark guess)
I particularly enjoy the absurdly undersized fixed metric fonts, text overlapping images making it even harder to read, fancy fonts that are blurry/hard to read and make scrolling painful, nonsensical/confusing heading orders making handheld/screen reader/keyboard navigation a PITA, fixed width layout that isn’t even 1024 friendly, lack of anything even resembling graceful degradation for images off or CSS off, over half a megabyte of javascript for NOTHING, 68k of CSS doing what’s probably more like 24k’s job, and a whopping 98k of markup (that doesn’t even mod-deflate/gzip well?!?) for 9.8k of content…
Which a 10:1 code to content ratio screams “go back and learn some html”.
Of course, it’s got jquery bloat from hell with that wonderful 667k of javascript (that again isn’t gzipping very well either – 479k compressed?) in 20 files. Combined with the 308k of images (FOR WHAT?!?) in 56 files it’s no wonder that on my 22mbps connection it takes 30 seconds of overhead on firstload. Nothing like blowing 997k after compression (1,249k uncompressed) on delivering 9.8k of plain-text… basically doing somewhere under 140k’s job.
Let’s peek under the hood (all of the above is before I even look at the source) – XML prolog, well that explains all the IE conditionals for nothing since you’re having to probably spend twice as much development time CSS hacking around being in quirks mode… Of course, it has the IE conditionals for nothing and at the pinnacle of ineptitude, it has to target EVERY VERSION of IE… SMOOTH. (not)
Naturally it gets to have the class on body for nothing, opens at least six tags before it even gets to the first actual content element, uses clearing DIV like it’s still 1998, has an invalid/inaccessible form, triple wraps form elements in div and span for christmas only knows what reason, has style inlined in the markup for no good reason other than bloat… all adding up to hundreds of div doing tens of DIV’s job. In fact, this suffers from one of the WORST cases of divvitus I’ve ever seen.
H1/H2 pairing doing h2/small’s job, first heading on the page a h3 – oh yeah, that’s a joy to navigate. Of course, you HAVE to have classes on EVERYTHING because you couldn’t possibly have semantic markup and make use of the “cascading” part of “cascading style sheets” – or as I like to call it “not every ejaculation deserves a name”.
I’d be shocked if the majority of people landing on that site even bother waiting for it to finish loading. Pretty much EVERYTHING I’ve come to expect from someone who says they don’t care about validation… still basically writing HTML 3.2 and slapping a tranny doctype on it as if that’s somehow ‘ok’. “I don’t care about validation” is usually just a surface symptom of deeper rooted problems with a coders attitude – which is reflected across that entire page.
Much less that even with the trickle of visits that mess must be getting, it’s probably killing your host in the process as well due to it’s RIDICULOUS page size… well, assuming anyone actually lets the page load.
Seriously, if this is your idea of a working website?
http://www.cutcodedown.com/for_others/transio/medicalMJ.jpg - 333k
I truly pity anyone DUMB ENOUGH to partake of your services. I’m ASSUMING what it looks like in FF is what you are trying to do – too bad it’s the ONLY browser that does so here.