The race site for the Surfer’s Path Marathon and Capitola Half Marathon has launched! Race day is May 20th, 2012.

Website Here:

http://events.lifestyleescapes.com/

The price of freedom is eternal vigilance.

~Thomas Jefferson

SOPA is bad for free speech, free expression, innovation, and the American sense of liberty. Censorship is not the way to stop piracy internet piracy.

Join the Protest:

https://www.google.com/landing/takeaction/

http://en.wikipedia.org/wiki/Stop_Online_Piracy_Act

http://americancensorship.org/

DiCarlo Arts has just rolled out a tablet version of the website, customized for view on the iPad, with a UI developed specifically for a touch device.

Just go to www.DiCarloArts.com on your iPad to view.

The Sphider search engine is a nice, compact program you can add to your site to provide a search function without having to go through Google or some other paid service. It offers all the features you’d expect from a search engine, including weighted searches, customizable output, and an easy-to-use administrator panel. The one thing it lacks, however, is the ability to search substrings.

Let’s say, for instance, you have a page on your site talking about an office you have in Europe. Searching for “Europe” comes up with the page without any problems, but if you type in “Euro”, you get nothing. For people used to using Google, this is unacceptable, because it tells the user that there are no pages with any version of “Euro”, including “European”.

The reason for this is twofold:

1. Sphider uses a series of tables in a database to organize and weight all the words on each page. To make the search faster, rather than one large database for each word, they are split into 17 databases, based on the leading character of the word when the MD5 hash of the string is calculated. For those of you unfamiliar with MD5, here is the official definition.

From RFC 1321 – The MD5 Message-Digest Algorithm:

The MD5 message-digest algorithm takes as input a message of arbitrary length and produces as output a 128-bit “fingerprint” or “message digest” of the input. The MD5 algorithm is intended for digital signature applications, where a large file must be “compressed” in a secure manner before being encrypted with a private (secret) key under a public-key cryptosystem such as RSA.

Just think of it as a function which takes a word and spits out a consistent string of gibberish. For Sphider’s purposes, the first digit of that string is the number of the table of words to look through. Since there are 17 possible values, that means that the search only has to look through one table, which is 1/17 of the total number of words. In essence, the search is faster.

The problem comes in when a term like “Euro’ puts out a different MD5 than “European’, which means that Sphider will try to look in a different table. Because of this, a search for “Euro’ will never come up with “European’.

2. The second reason involves the query that Sphider sends to the database. Here it is:

$query1 = "SELECT link_id from ".$mysql_table_prefix."link_keyword$wordmd5, ".$mysql_table_prefix."keywords where ".$mysql_table_prefix."link_keyword$wordmd5.keyword_id= ".$mysql_table_prefix."keywords.keyword_id and keyword='$searchword'";

The $wordmd5 portion is the MD5 code calculated earlier, which tells which table to search for that particular word. What you’ll notice, though, is the portion for keyword=’$searchword’. Since the search term and word in our example aren’t equal, this will never turn up a correct result. What we need to do is alter the equation to not only find things that are exactly equal, but are like each other. The new portion of the code we end up with is:

keyword LIKE '%".$searchword."%'

The LIKE operator is MySQL code for a simple pattern matching operation, and the % characters indicate that there can be any number of other characters before and after the search word. So, “asdfasdfeuro”, “euroasfasdf”, and “asdfasdfeuroasdf;alk” would all be acceptible.

Now that the basic search query is altered to match variations of the search term, and not just the search term exactly, we have to alter it further so that it searches every table of keywords, because those other words which include the keyword as a substring might have a different MD5 code, and thus be in a different table.

This is where the hack gets ugly.

Yes, there are longer ways of organizing things that would provide a more elegant solution, but the quickest, simplest solution is to simply make a query for each table, and combine them all with the UNION MySQL term into one big query. Hold your breath:

$query1 = "(SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword0, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword0.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword1, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword1.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword2, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword2.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword3, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword3.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword4, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword4.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword5, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword5.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword6, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword6.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword7, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword7.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword8, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword8.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyword9, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyword9.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyworda, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyworda.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keywordb, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keywordb.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keywordc, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keywordc.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keywordd, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keywordd.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keyworde, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keyworde.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"UNION (SELECT distinct link_id, weight, domain from ".$mysql_table_prefix."link_keywordf, ".$mysql_table_prefix."keywords WHERE ".$mysql_table_prefix."link_keywordf.keyword_id = ".$mysql_table_prefix."keywords.keyword_id and keyword LIKE '%".$searchword."%')".
"order by weight";

What this does is disregard the MD5 code for the search term, and search every table anyways…even the ones Sphider thinks shouldn’t have the search term in it. While this does theoretically slow the search down, if the site isn’t hundreds of pages long, the search time won’t be very much longer. What is does, is turn up results for “European” when someone searches for “Euro”, which is well worth the sacrifice of a few tenths of a second in search time.

The main website has just undergone a complete series of updates. All the Flash pages have been converted into more accessible web formats, constructed with traditional HTML, CSS and JavaScript. I believe this gives a much better user experience, and makes the information accessible to those without Flash.

The design of the gallery is meant to give the viewer a totally interactive experience, and indeed, there are many, many functions available throughout the page for the user to find as they navigate the material. Some are obvious, others are easter eggs, hidden for the user to find. Images are dynamically loaded via JSON data, which means the same template can be used to power many different galleries, and the information can be updated from anywhere without the need for accessing a MySQL database. The gallery functions less as a traditional webpage, and more as a web “app” making full use of the entire browser window.

The site relies heavily on jQuery, a free, open-source JavaScript library which powers much of the site’s interactivity and animations. The site also makes use of alpha-layer png files to add depth and subtlety to the interface.

The main block of thumbnails, once loaded, respond to the user’s mouse movements, allowing them to scroll through a potentially high volume of images by moving through a hot-spot located in the center of the page.

As always, the site is in a continual state of renovation, and as more material is added to the new framework, there will no doubt be more changes and additions. Keep in touch!

The new mobile version of DiCarloArts.com is online. Visit www.DiCarloArts.com on your iPhone to be redirected and experience the new site!

More info to come…

In a continuing series of free stock artwork, here is a set of vector images of ballet dancers in different poses.

Download Compressed File Here

DancerSilhouettes.zip (includes .ai, .eps, and hi-res .jpeg files).

You are welcome to use these images royalty-free, so long as you don’t a) pass them off as your own work, b) distribute them without redirecting people back here, or c) sell them for a profit.