Does Google care about ALT attributes? “it seems to be a little schizo on this – there are times it cares and times it doesn’t”
“a whole industry has sprung up trying to figure out what Google likes and dislikes.”
An industry? Are there legal ways to improve your Google status?
“One of the biggest arguments is over the question of whether Google likes/doesn’t care about web pages that use valid X/HTML code.”
What is the dope on this?
Finally, this last question came from a different article:
Dude, hate to say it but Jakob Nielsen is long gone. Usability, having become a science (with a very simple solution) has, resulted all major company web sites based off of half a dozen templates and the only real differentiator now is the search function. Enter KM.
Of course, no one can figure out how to make Knowledge Management useful/usable.
Most websites now suffer from bloat, same way software that’s been around for 10 years does. I just ordered something off of Amazon for the first time in 5 or 6 years; what a pain in the ass.
I don’t know who web designers are making their sites for: users or other web designers.
Of course, my own site is completely nonnavigatable, but it was designed like that.
“Legal”? Of course it’s legal. You might tick off Google and get your sites completely unindexed (it’s happened quite often), but certainly nothing illegal about it. And it is an industry in that there are companies selling their services to do just that. And they have a lot of buyers.
You have to create a system of “trustworthy” links to your sites. The more links that Google likes, the higher your ranking (in general). Some people have gamed this quite well. Then Google figures out how they did it, modify their algorithms, the game starts over.
I was searching for a once common product last summer. Something like 18 out of the first 20 hits were to strangely quite similar sites. They are really just one site with “corporate partners” (yeah, right).
It’s one thing to try and get your site listed high. It’s a whole 'nother piece of sleaze to get only your site listed high. This is basically spamming Google.
There are always people willing to make a buck ruining a good thing.
As to the web design issue. It’s like programming but worse. In programming, 1 in 100 programmers are actually good at it. The rest just get in the way. But with web design it’s worse. Really awful designers don’t even get fired. So less than 1 in 1000 are good. The quotes regarding Lynx are a fine example of people who shouldn’t be allowed to design web pages. (I constantly run into web pages now that work on only one version of MSIE! Now that’s double stupidity.)
The bottom line, almost every web page out there sucks. One reason Google has done so well is that they did their start page right. Hardly anyone has imitated that philosophy.
Alright, let’s see what we’ve got with this start page… Missing DocType, so it can’t be validated, a <style> tag with no type attribute, a <script> tag with no type attribute, color markup in the <body> tag, a <center> tag, a couple of layout tables, elements with their id attribute set to illegal values (like “1”), use of the <font> tag for setting color and size, and a lot of nonbreaking spaces for layout. All in under 4kb.
Not too bad, but they could do better, I think.
Not to be too much of a Google-worshipper, but it really does seem like the single best tip for making your site show up high on a Google search for whatever is “Make your page one that people who search for whatever will want to find”.
As for design, I personally think that your webpage looking good in Lynx is only as important as its looking good in IE, or Mozilla, or any browser. I don’t think that you should design with browsers in mind, though, as the author seems to be suggesting.
Looks like you answered that question yourself, buddy. I read that article a long time ago and it just confirmed my own beliefs. I don’t really care how many clicks it takes if I get there, and if I can do it without getting lost or being unsure where I’m going.
I’d much rather face a well-organized, multistep hierarchy where I can identify where to go at each step than either try to search through a list of dozens of links or try to determine which irrelevant category I should be looking in.
How many clicks it takes depends on the amount of information a site provides. If there are a thousand pages on different topics, then I don’t want to have to search through a hundred links on each page just so they can fit it into three clicks. On the other hand, no one likes to have to search constantly, deeper and deeper, for something that should be accessible.
I think the three clicks rule is a nice idea - but making everything nice and findable is the real goal. Three clicks can be a useful rule of thumb, but anyone who believed people always stop after three clicks has never monitored their own surfing. To bad it still seems to show up in a lot of web designs.
[nitpick]
With a thousand pages, you just need 10 links per page to satisfy the 3 clicks rule. Top page has links to 10 categories; each category has links to 10 sub-categories, each sub-category has links to 10 documents. 10x10x10=1000.
[/nitpick]
Yeah, I realized after I posted that expontential increases might dilute my point - but it’s still not unusual for me to run into websites that have far too many links on a single page, seemingly in order to require less clicking. Perhaps these sites have a lot more than a thousand individual pages.
