Would you buy a used website from this company?

Misaligned form button overlaying input

I’m a sucker for mechanical validation (in both senses), so when Gadgetopia linked to Silktide’s Sitescore website testing application, I hied myself hence.

Um. The very first thing I see, since I’m looking for it, is a grossly oversized input for your website URL, with, in Firefox 1.0.x, 1.5rc3 and current trunk, the submit button displayed too low and too far left, overlapping the end of the input. I don’t know or care (yet) whether that’s a Firefox bug or not: if you are selling website development, you should not look obviously broken in Firefox, not in the tail end of 2005.

Then, the very first item in the Detailed Breakdown, where I get dinged as “Very Poor,” is

None of the tested webpages made use of meta tags. This website is likely to be either very old or not well designed. Meta tags help promote your web pages to search engines.

In what century? Is this actually a ghost page, left over from the mid-nineties? You praise me for my “Popularity Rating” and “Popularity on Google” despite the fact that I’m currently in the sandbox, and you think you can teach me something about using meta tags?

I could push my score above the current 8.0 by testing while my first post is something right up my alley (they search for what they guess to be important keywords, and having identified “xml spec” and “xml parser” they didn’t find me, though amazingly enough I’m number 45 for “conforming xml”) and using some pretty pictures to make them happy, but there’s no point: I’m not going to do keyword-stuffed meta “tags” and they aren’t going to love me.

15 Comments

Comment by Breyten #
2005-11-21 01:42:58

I’m not sure what steroids they’re on, but apparently my site uses no tables for layout ;-) Makes you wonder what that table tag does then…

Comment by Phil Ringnalda #
2005-11-21 02:52:56

They probably got so nervous when they saw the warning not to remove the template designer’s credit because WE CHECK DAILY! that they were afraid to tell you anything bad.

WE CHECK DAILY!

That better be scripted, or an empty threat, or someone’s badly in need of more to do.

 
 
Comment by Dave Child #
2005-11-21 02:31:59

They also appear to have a rather poor Company Name Identification section. It assumes that the company name is always at the start of the URL (which is sometimes true, but by no means always) and makes no comparison between the URL and front page title, which would seem an obvious way to identify a company name.

On the subject of meta tags: while they are not as powerful as they once were, they are still useful in many ways. Google sometimes uses the meta description instead of a page snippet in results, allowing you to occasionally control what is displayed in that part of the results. The meta keywords tag can also be useful when well used, especially with mis-spellings, if I recall correctly.

I’d be interested to know how they distinguish ”quality links” from normal links as well…

Comment by Phil Ringnalda #
2005-11-21 02:59:56

Yeah, I remember reading that you can slip in a meta description, but what are the rules? Something like the search phrase has to appear in the meta element and in body text? As to keywords, I’m pretty sure that I’ve read (though unless it’s Matt Cutts writing, you don’t ever know for sure) that putting keywords in a meta that don’t also appear in the body text serves just one purpose: making their spamtenna prick up, as they suddenly develop a sincere interest in every single other thing you do.

And for a weblog, having a description, especially the sort we generally use (”a digital magpie”, feh) is vastly less useful to the searcher than a ransom note excerpt, which might actually tell them that their two search terms only appeared in two wildly separated sentences. I clearly confuse and disappoint enough searchers as it is, without trying to lure in more with a clever and ambiguous paragraph of description to hide my sins.

Comment by Aristotle Pagaltzis #
2005-11-21 03:40:26

a ransom note excerpt

Aphorism, or typo?

Comment by Phil Ringnalda #
2005-11-21 03:51:02

Term of art, I hope, unless I’m just not seeing a typo in it. It is almost 4am, but I’m pretty sure that’s the correct term for an excerpt made up of sentence fragments ripped out of their context.

 
 
 
Comment by Phil Ringnalda #
2005-11-21 03:36:25

Forget your company name, you’re number four for ”cheat sheet”? I’m impressed.

I was puzzled about the link report, too: I’m in the sandbox, link:weblog.philringnalda.com on Google returns no results, and yet I have 1970 links, none of them ”quality”? Maybe some combination of terms on Yahoo, which doesn’t hate me as much right now, but I couldn’t find one that gets the same number.

 
 
Comment by Aristotle Pagaltzis #
2005-11-21 03:53:32

They ding me with a score of about 3 on “marketing,” but give me close to 10 on other counts, so I come out with a score of 6.8.

Apparently, http://plasmasturm.org/ carries a company name of “Silktide.” The significant phrases purportedly all include “web design,” and several of them contain “nottingham,” “leicester,” “derby” and “uk.” And there is a search facility. But noone is linking to me, they say.

What the…?

Did their crawler actually visit plasmasturm.org – and is it the one in this universe?

Comment by Aristotle Pagaltzis #
2005-11-21 04:07:35

Okay, mystery solved – I sent it to plamasturm.org (observe the “s” missing from the “plasma” part).

Which makes me wonder, since that domain does not even resolve: why did the crawler not complain, and what did it actually test?

Run against the correct address, the results are less surprising. I am amused that it complains about the lack of forms and images. And apparently, the minimum level of profanity you can achieve, other than none at all, is “moderate,” because there is exactly one instance of the F word on the entire site and not a single other bad word, and it dings me about that.

What intrigues me is that it finds me in violation of accessibility, because the validator complains about the @xml:space in my weblog archives (served as The Real XHTML Thing). Am I in error, or the validator?

Comment by Phil Ringnalda #
2005-11-21 20:40:35

That’s the sort of thing I tend to get wrong, but I think the validator is correct, and you are in error: xml:space must be declared if you are going to validate against a DTD, and XHTML 1.0 Strict’s DTD only declares it for style, script and pre.

Of course, using an (X)HTML validator as a proxy for accessibility is somewhere between a huge stretch and insane, though the one error it pointed out to me was images without alt attributes, so in that case it did work.

 
 
 
Comment by o-juice #
2005-11-24 17:10:33

If you’re interested enough to see some more rather puzzling and silly ”results” from validating your site through them, create an account (via their tools section, log in and increase the number of pages (25 is the maximum) the tool crawls through and make sure you tick the ”Show technical results?” checkbox.

 
Comment by Jesse Ruderman #
2005-12-03 16:45:31

Another sitescore bug: when I enter ”www.squarefree.com/pornzilla/”, it chops off the last slash before requesting the web site, and then takes off points because the URL redirects. I can’t imagine what they were expecting.

At least they stopped taking off ~7 points from each category (marketing, design, accessibility, experience) when a site mentions porn, like they used to.

 
Comment by Arve #
2005-12-03 18:34:02

Wow. I cannot quite find the words needed to describe what I feel:

  • It frequently fails to display a progress indicator, so I’ve closed the browser window three times, only then to reload it all again. Is there a home for abused web servers?
  • On checking Anne van Kesterens site, it makes some ludicrous claims about accessibility
  • When testing 25 pages, and asking for a detailed report, my site ranks as ”Excellent” in all but one category; Images. Yet it only rewards me with 8.7 (or 8.5 in the image they so bad want me to link to)
  • If you log in, the front page is actually missing the form element that allows you to change the depth of their ”analysis”
  • I’m now running analysis on a second site I run, I started that analysis before reopening this window, locating the form field, made typos, corrected typos, made more typos, rechecked the window. Still no progress indicator.
  • If you register, they never send you a confirmation mail to confirm your account, so you’re pretty much free to sign up with a bogus name and address, as long as you’re able to remember your passwords (or have some device or program remember it for you)
  • Their in-beta tool, named GoogleWatch seems to be a walkin’ talkin’ violation of any Terms of Service Google has — you’re not supposed to do automated querys against google)
  • Oh, did I mention: I think their ”querying 25 documents” failed. I just restarted it.

I guess I could go on, but I think I’ll just forget this tool exists instead. My life will be so much better.

 
Trackback by Software Guide #
2006-01-09 12:51:58

Programmatische Prüfung der Webseitenqualität

Silktide.com bietet mit Sitescore ein Online-Werkzeug zur Überprüfung der Webseitenqualität.
Die Prüfung von sw-guide.de ergab u.a. folgendes Ergebnis:

Dies wird im Detail heruntergebrochen wie etwa:

Google-Popularität
Popularität weltweit
Feat…

 
2006-01-22 10:48:38

[…] Auch bei Wolfgang Sommergut oder Phil Ringnalda kann man nachlesen, wie sie diesen Service beurteilen. Kategorie: Web-Anwendungen | PermaLink | Trackback | Abonnieren […]

 
Name (required)
E-mail (required - never shown publicly)
URI
Your Comment (smaller size | larger size)
You may use <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <code> <del datetime="" cite=""> <dd> <dl> <dt> <em> <i> <ins datetime="" cite=""> <kbd> <li> <ol> <p> <pre> <q cite=""> <samp> <strong> <sub> <sup> <ul> in your comment.