PHP - Easy Reflections Script

Core PHP has a cool little script for making easy reflections that will take any image and create a reflection of it. Options include caching, height of the reflection, tinting, and fade start/end.

While this is something easily accomplished in something like Photoshop, it could come in handy to automatically display reflections for user-uploaded images in a public gallery site for instance.

Just Say No to Nofollow

I disappear from the Web for a couple of years and come back to this Nofollow thing from Tom Christensen.

"... a lot of SEO types were posting about nofollow again. The new twist is they’re trying all sorts of plugins and gadgets to selectively pass or bar following links from their blogs for PR.

People, this is getting really old. And really stupid. Just turn the damn thing off already."

I am going to have to agree with Tom on this. It is really stupid.

Also, the whole Wikipedia decision to add Nofollow to outbound links is stupid. And Andy Beal's campaign to Nofollow Wikipedia is stupid, even though it looks like he is Following their lead when it comes to trackbacks and comment links. Not quite calling the kettle black, but close to a very dark grey.

Carsten Cumbrowski opines:

The hope is that the return for spamming Wikipedia will be so low that it does not even make any sense for those spammers that don’t need much return to be happy.

You can live perfectly fine in India for $1 a day for example. If Spamming Wikipedia reduces that down to $0.25, the spammer will probably look for other targets. And those other targets will also go away eventually, but that is a complete different story.

Well isn't that special! Keywords here are "other targets" (eg: You, Me and Dupree). And just how will those other targets eventually go away?

And why is it a completely different story? It is the story. Wikipedia effectively kicked the spammers out of their yard and they are coming to a neighborhood near you. Is it just me, but how does this combat spam?

Its okay to follow NoFollow. Follow?

When Google first suggested Nofollow back in 2005, it didn't take long for a spec to be drafted up by Technorati. MSNSearch and Yahoo!Search jumped quickly on board supposedly, along with scores of blog software developers and proponents.

The spec abstract has nothing to do with not following a link, just applying no weight to the link itself:

By adding rel="nofollow" to a hyperlink, a page indicates that the destination of that hyperlink SHOULD NOT be afforded any additional weight or ranking by user agents which perform link analysis upon web pages (e.g. search engines). Typical use cases include links created by 3rd party commenters on blogs, or links the author wishes to point to, but avoid endorsing.

When the big three of search come out and say "they will respect Nofollow", what does that mean really? After reading dozens of comments to this regard, I am under the impression that spiders can and will follow the links if they so choose ... they just will not apply any weight to the link. A link is a still a link then.

Is SPAM the cholera epedemic of the Internet?

Should we burn down an entire village of Wikipedia huts just to eradicate a plague, only to have that plague show up in somebody elses village? Where does it stop? I think with all the Doctors and Chemists at Google General, they can come up with a better cure than this.

My hat goes off to Slashdot who use heuristics, karma and other factors in combination with nofollow to combat spam. Also, to Blogoscoped for their "fading nofollow" policy. And to other like minded bloggers who are just saying NO to NoFollow.

This is the kind of responsible forward thinking that we need to be doing, not to mention doing the SE's job for them to boot. It seems to me that any SE should have the ability to differentiate between a blog post and a comment and not apply too much weight to the comment anyway. Why should we have to tell them this?

Server Farms: Cheap Electricity Key Issue

Server Farms are cropping up across the country. Google has a few farms along the banks of the Columbia River in The Dalles, Oregon. Upstream in Quincy, Microsoft has opened its first of six planned server farms, a mammoth 474,000 square feet and surrounded on three sides by fields of potatoes, beans and broccoli. The mighty Columbia has been a source of cheap electricity for decades and it isn't any wonder why this area was chosen.

Meanwhile, Microsoft and Cisco are investigating the possibility of establishing server farms in Iceland powered fully by renewable energy. The power comes from both geothermal and hydroelectric sources and is so cheap that in the wintertime some sidewalks in Reykjavík and Akureyri are heated.

Google is also building a massive server farm near Eemshaven in the Netherlands, where 100,000 servers will have access to 30 megawatts of power, some of which will be delivered by windmills.

And what if you cant find cheap electricity? Then you have the Iowa State Senate pass a bill that gave Google a tax break on the sales tax from utility bills and a property tax break for the site of the center itself.

There is not just the cost of electricity that is an issue here. Larry Page cares about the polluting effect that Google may have. Apparantly the Internet is not a very clean industry, and mega-coms like eBay, Amazon, Yahoo, Microsoft, Google are partly responsible for global warming for the server farms are hardly "carbon neutral". Here we go again ... Google is evil, bad Google.

Semantic Markup: Is it worh it?

After writing a simple page such as About, I ran it through the W3C Markup Validator with the Show Outline option ticked to get their slant on the page.



Everything is kosher up until the outline for Graphic Software, then the sidebar headings bled into it. The headings for the sidebar are just general placeholders for menu text to be defined later (as the site progesses).

After seeing this, there seems to be two courses of action to take. One is the easiest, the other is more progmatic.

The easy approach

Easily enough, I can simply change the heading of the sidebar to an H2. This would put the lists in that area within their own 'section' and seperate from the post itself.

The downside is that every post that I make will have this sidebar 'attached' to it in some way. I am not sure if I want that attachment, but semantically it does make some sense.

The harder approach

Since the sidebar is set up as Php include, I can perform a little magic with it. For example there can be 'targetted' topical links included in the sidebar.

This would require a little more effort on my part, but one that I am sure that I can overcome.

I do know that Blogger will publish seperate "tag lists" for the blog posts. How you can discern them from the regular RSS feed is questionable. You would have to take the published Html pages and convert them into an actual feed.

My Sidebar is already set up for the challenge. It is adaptable. It only needs a little encouragement.

CSS Techniques you cannot Live Without

IsPopularOnline brings us 53 CSS Techniques you cannot Live Without a compilation of links with thumbnailed representations of each. The list represents the best written articles about CSS that the Internet has to offer and presents them in an eye-catching way. Very sweet one to bookmark.

ALA Web Design Survey, 2007

ALA is sponsoring The Web Design Survey, 2007, and yes I just took it.

"People who make websites have been at it for more than a dozen years, yet almost nothing is known, statistically, about our profession. Take the survey and change all that."

One sidenote, ALA admits that the survey does not validate due to a glitch in the software used to generate it with. They are working on it.

Semantic Web Markup for Blogs

During the course of revamping the Webstractions site, I came across a well thought out article by Chris Pearson on semantic markup for blogs. The article was spawned over the idea of how an open-source blog software such as WordPress, MovableType, or TypePad is virtually void of truly semantic templated themes.

Semantics aside, the other thrust of good markup implied that it will garner you with better SERP placement. Outside of the fact that it is 'Google recommended', there is no tangible proof that properly marked up pages will gain you more advantage. Having said that however, I am not one to tempt fate and we all must strive to eek out every bit of edge we can.

Blog Comments & Trackbacks

Blogs are strange animals by nature and really do not conform to an atypical 'content' page. In fact, they are very similar to Forum sites. Start your post off with a strong title and 250 words of keyword rich content around that title, then hope for the best. By the time you get 30, 40 to a kazillion or more comments following your post, then it pretty much gets watered down in the process.

Blog comments are a necessary evil. On one hand, they will make the spider go crazy if the blog is busy with them. Nothing like a post that is constantly growing with comments. On the other hand, the quality of comments can degrade the post and dilute it with tons of 'good post, mate' and 'this helped a bunch, dude' types of entries.

Outbound links from comments can be more degrading than the comment itself. It is rare that those links go to a like-minded site, or have anything to do with the original post.

Trackback linking is very good to have. It is what makes the blog an almighty spider delight.

I would delegate each of these to their own DIV with one heading tag at the top of each. H2 or H3 for the trackbacks, and comments lower down the chain with an H4 or H5 even.

I noticed that Chris was surrounding the outbound comment links with their own H3 elements, this is not right in my opionion. Too much emphasis is being put on non-supportive no-follow link and the commentors name.

I question whether individual comments are worthy of a heading element in the first place. A more elegant solution would be the use of a definition list as per this W3C Recommendation:

Another application of DL, for example, is for marking up dialogues, with each DT naming a speaker, and each DD containing his or her words.

My View on Sidebar Markup

Chris was spot on about proper usage of <title> and <h1> elements, and even using sub-elemental <h2> or <h3> within the posts. The use of them in the sidebar, however, may be open for interpretation and requires a little more thought.

I am (currently) using one <h3> element and short description at the top of my sidebar. The areas for Latest Posts, Syndication, Archives, and Powered By are all served up within their own definition lists (DL). My slant on this, right or wrong, is that they are all groups of lists and as such, each list needs a definition.

In Conclusion
Strong TITLE and H1 are key. Add in an H2 or two within your post as needed. Delegate one heading element H3 or lower to each section, listed in order of importance: Trackbacks, Sidebar, then Comments.

Trackbacks, Latest Posts, Blogrolls, Syndication, etc are all lists and should be treated as such. I chose to use DL to contain mine. I don't think it really matters if you use UL instead, this is just my preference and doing so eliminates the use of more Heading elements to define them.

Comments, per the W3C, qualify as a Definition List (DL). In my opinion, I think this is a little quirky and I will have to give this a little more thought. Any opinions would be greatly appreciated.

New Look & Direction for WebStractions

After a 2 and half year hiatus, I am back. Wondering if anybody out there missed me, of course you did.

Anyhow, the site has a new design and direction. Right now I am picking up the pieces of setting up a new host and recovering most of the blog (for archival purposes)

Will be back with more.