In 1989 we had zero web browsers as we know them today, allthough just about to be invented around the corner. In 1999 we had two web browsers fighting a death match, Internet Explorer and Netscape Navigator - a fight with Netscape cleverly lost by dying and coming back several open source reincarnations of which Firefox of course is the most well known today. 2009 is turning out to be yet another battle year for browsers, this time many more of them! We have (in no special order) the newcommer Google Chrome fighting Firefox and Internet Explorer (mainly the PC-side). We have Opera who has cut out a piece of the action on several systems but shine mostly in portable devices. Safari is ruling the Macintosh but is starting to get some interference from Firefox.

Well that is now, what is next? I ready a post about current state of browser development, and many of the major browsers have a beta our that will maybe go live sometime during the next year. While this might be very good news for home users I am sure it will mean alot of work for someone like myself who create on-line applications. There used to be a lot of tuning to make web pages and applications look and work the same on the old “two major browsers”, now we have at least 5! Unless the browser developers makes a great effort to follow the rules of the standards each web page have to compensate for how a particular browser parses the data.

In the past Internet Explorer have seemingly intentionally ignored several standards in favour of making programmers like myself forced to make pages look good on their browser. Internet Explorer is afterall the dominating browser and it have to work. The question is if this strategy is allowed to continue. I really hope for the sake of us programmers that while there are five new browser versions about to be released that several of them will render the basic pages using the same rulset.

Linux is a free stable and resourcefull operating system yet to prove it’s full potential as a desktop operating system. While the greater part of the market is still Windows based the introduction of the UMPC (Ultra Mobile PC) have turned things upside down for Microsoft. Instead of buying very expensive hardware and install Vista many choose to buy less hardware and get either a Linux distribution or (even worse?) a several years old Windows XP system.  Microsoft was not prepared for this turn of events and have been forced to extend the lifespan of their XP-series which they where planning to phase out in favour of Vista.

This change in consumer behaviour to “not buy the best” anymore, have favoured Linux who with their versatile distributions have a stable modern operating system for just about any new or old hardware out there. My five year old laptop which where being exhausted by running Windows XP in combination with a virus tool runs perfectly smooth with the Ubuntu distribution (well I could theoritically run XP without a virus tool but seeing I travel alot with it that would just be beging for it to crash with all the unsecure networks I connect to).

With the UMPC Linux have gained a small part of the desktop market, but what is holding it back to take an even larger part? Is it the hardware support? Well granted some hardware vendors only supply drivers for Windows systems, but most hardware have at least decent support in Linux. Is it Internet related software? Hardly, Firefox is arguably better than the current version of Internet Explorer (granted Firefox is available on both systems) and also e-mail clients like Thunderbird are just as easy as Outlook Express to setup. Are the office tools sub standard? Open Office and GIMP make a very good competition for their commercial counterparts Microsoft Office and Adobe Photoshop, so except for prior experience in one system there really is no major difference for the home desktop user there. Is it the ease of use? Again, trying out Vista will be just as confusing for a XP user as it would be switching to Ubuntu. Most things have changed places anyway. Also the update feature in Ubuntu is much better than Windows update in my opinion. The ability to upgrade all installed applications through a common interface is very easy to handle for an inexperienced user.

Is there any major reason to hold back? Well… I have to say yes. There is one single left over reason that I see quoted time and time again: “I use Windows XP becuase I want to play games”. Almost no PC game whatsoever have Linux-support. There are, often technical, HOWTOs on how to get this or that game running under Linux but it usually involves pages of instructions. If we have everything else, surely games can’t make a major difference for a desktop operating system? Again, I’d have to say yes it does. Why? Simple, the kid in the family who usually is the computer wiz is the one who helps all his family and friends install their computers. If this kid runs XP back home to play computer games on he will install XP on all other machines, because this operating system he knows and can “support”. Furthermore if all else being equal, if there is one major drawback in a single area, whoever wins that will probably win the battle.

Windows XP have DirectX which is supported and used by I would dare say every game manufacturer for the PC market. Linux have no equivalent interface. Linux have “emulated” Windows and even tried to port games individually but the major all-supported interface is not there. Game companys can’t make a release for every distribution out there (since many are so different it would be required), but if a common interface for all distributions could be established that would increase the chances of Linux game releases.

In my opinion, this is one of the major reasons Linux is still held back as a desktop operating system. Far from the only reason, but a major one.

It’s been a while since I had time to post some useful information here, meanwhile so you don’t turn into zombies or even worse get attacked by them here is some helpfull code from Google themselves:

http://www.google.com/robots.txt

And, just in case they change the file, here are the top lines from their robots.txt:

User-agent: zombies
Disallow: /brains

User-agent: *
Allow: /searchhistory/
Disallow: /news?output=xhtml&
Allow: /news?output=xhtml
Disallow: /search
...

Today I’ve improved the starting page of this domain, www.fireflake.com. It looked very booring and I thought I’d have to do something with it but at the same time I knew I was to lazy to acctually maintain some interesting subjects there as well. The solution was simple! Use my blogs to feed the information to the starting page! Here are some of my php and sql code used, in case you want to try something similar yourself.

First of all, if you’ve never done any PHP before, do not fear, it’s super easy! Without explaining it, here are my open and close database files (I keep them seperate since they will most likely be included in many pages):

open_db.php:

<?php
$dbhost = ‘localhost’;
$dbuser = ‘username’;
$dbpass = ‘password’;

$conn = mysql_connect($dbhost, $dbuser, $dbpass) or die                      (’Error connecting to mysql’);

$dbname = ‘dbname’;
mysql_select_db($dbname);
?>

close_db.php:

<?php
mysql_close($conn);
?>

Just modify these files with the values for your database and you are ready to do some mysql-powered-php-scripting! Next simply include these files in the file where you want to use the database, in my example index.php:

<!DOCTYPE html PUBLIC “-//W3C//DTD HTML 4.01 Strict//EN” “http://www.w3.org/TR/html4/strict.dtd”>
<?php include ‘open_db.php’; ?>
<html>
…
</html>
<?php include ‘close_db.php’; ?>

Now lets make the MySQL database print something to the index-page! Here is an example:

<?php
$query  = “select now() as mysqltime”;
$result = mysql_query($query);
while($row = mysql_fetch_array($result, MYSQL_ASSOC))
{
echo $row['mysqltime'];
}
?>

This will print a timestring from your MySQL. I used this as an example just because it should work everywhere. Now simply substitute the query with something you want from your database and the printout inside the loop with what (and how) you want it printed!

To take a Wordpress example, I want to use the tags/categories as keywords for the META-tag of my start page. I use a cleaverly written SQL query that will give me all the unique keywords for all my three blogs sorted in order of usage:

<?php
$query  = “SELECT name,`count` FROM tech_terms t join tech_term_taxonomy tt on (t.term_id = tt.term_id)
union distinct
SELECT name,`count` FROM game_terms t join game_term_taxonomy tt on (t.term_id = tt.term_id)
union distinct
SELECT name,`count` FROM wp_terms t join wp_term_taxonomy tt on (t.term_id = tt.term_id)
order by `count` desc limit 0,50;”;
$result = mysql_query($query);
$keywords = array();
while($row = mysql_fetch_array($result, MYSQL_ASSOC)){
array_push($keywords,$row['name']);
}
$keywords = array_unique($keywords);
foreach ($keywords as $key){
$meta_key_string .= $key.”,”;
}
?>
<meta name=”keywords” content=”<?php echo $meta_key_string ?>fireflake”>

This will make my home meta tag always up to date with what I write about! The keywords will also be listed in order of relevance since they are ordered by “count” from the database.

Another great little script takes out the latest posts in a blog and prints them (and links it to the main article):

<?php
$query  = “select * from tech_posts where post_status = ‘publish’ order by post_date desc limit 0,2″;
$result = mysql_query($query);
while($row = mysql_fetch_array($result, MYSQL_ASSOC))
{
echo “<h3><a href=\”" . $row['guid'] . “\”>” . $row['post_title'] .”</a></h3>”;
echo “<p>” . str_replace(”\n”,”<br />\n”,$row['post_content']) .”</p><br clear=\”all\”>”;
}
?>

There are probably lots of fun you can have with the Wordpress database, it’s very simple and easy to learn so it’s very easy to start writing code like this!

Hope you found something useful here!

Marking the three year anniversary of the European Parliaments no to software patents the 24th of September is being made “Stop Software Patents World Day”.

On “Ask Slashdot” the readers recently got to answer which their favourite programming reference sites where for different languages. While most of the gems that where posted are quite well known I thought I’d make a quick summery of all the valuable links posted there.

Most languages have recieved at least one link but it’s interesting to see some obscure languages in there while others are not even listed (Visual Basic? educational languages like SmallTalk? and at the time of this writing no reference to either MSDN or Google Code). It’s a good group of links though, many informative pages here for anyone who uses these languages.

These are all the sites that programmers recommended (who else visits slashdot?) in alphabetical order.

Algol
http://www.algol68.org/

ASP
http://www.asp.net/

C
http://cprogramming.com/
http://www.acm.uiuc.edu/webmonkeys/book/c_guide/
http://www.devshed.com/

C++
http://www.cplusplus.com/
http://www.devshed.com/
http://www.parashift.com/c++-faq-lite/
http://www.sgi.com/tech/stl/

CSS
http://w3schools.com/css/default.asp

FORTH
http://www.phact.org/e/forth.htm

HTML
http://w3schools.com/html/default.asp

Java
http://java.sun.com/
http://java.sun.com/javase/reference/api.jsp

JavaScript
http://developer.mozilla.org/
http://w3schools.com/js/default.asp

Logo
http://el.media.mit.edu/Logo-foundation/logo/programming.html

LUA
http://lua-users.org/wiki/
http://www.lua.org/
http://www.lua.org/manual/5.1/

Lucid
http://www.haskell.org/haskellwiki/Lucid

Pascal
http://www.freepascal.org/docs-html/ref/ref.html

Perl
http://cpan.org/
http://use.perl.org/
http://www.perl.com/
http://www.perlmonks.org/

PHP
http://php.net/

PL/I
http://www.users.bigpond.com/robin_v/resource.htm

Prolog
http://www.logic.at/prolog/faq/faq.html

Ruby
http://ruby-doc.org/core/
http://api.rubyonrails.org/

Python
http://code.activestate.com/recipes/langs/python/
http://python.org/

Scheme (LISP)
http://srfi.schemers.org/
http://www.schemers.org/Documents/Standards/R5RS/

SNOBOL
http://www.snobol4.org/

TCL
http://en.wikibooks.org/wiki/Programming:Tcl
http://tcl.tk/

Multilanguage
http://c2.com/cgi-bin/wiki
http://www.gotapi.com/html
http://www.quickref.org/
http://www.regular-expressions.info/
http://www.rosettacode.org/wiki/Main_Page

Not purely programming but related
http://en.wikibooks.org/wiki/Main_Page
http://mindprod.com/jgloss/unmainprinciples.html
http://www.google.com/ (!)

Not mentioned links, added by me
Google Code
Visual Basic
SmallTalk
MSDN

I must start off with admiting that my PHP skills are very limited, however being a very experienced Perl hacker this is very similar to me.

I needed a script that checked for a normal HTTP response from another server, a status script to see if the other server(s) where behaving as they should.

The resources on-line for PHP are great and I quickly found the code needed to retrieve a remote page and read the contents. I also found some tutorials describing how to use this code. My version ended up looking like this (thanks phptoys.com for the tutorial!):

<?php
// check responsetime for a webbserver
function pingDomain($domain){
    $starttime = microtime(true);
    $file      = fsockopen ($domain, 80, $errno, $errstr, 10);
    $stoptime  = microtime(true);
    $status    = 0;

    if (!$file){
        $status = -1;  // Site is down
    }
    else{
        fclose($file);
        $status = ($stoptime - $starttime) * 1000;
        $status = floor($status);
    }
    return $status;
}
?>

This code works just fine as long as the server is up, but fsockopen generates an error message when the server is down. Here is the problem: the code is intended to verify that the server is up and take action when it is not and therefore I do not want to display the error message (as I half expect it).

I’ve tried to find a way to supress this error message (that I know will occur at some point) but could find no on-line resource describing it. In lack of a proper solution I instead made a hack innovative solution. I know when the error message will show up and that it will be written to output, to avoid this I simply put the function call in an “invisible” space of HTML:

<div style="display:none;">
<?php
$status = pingDomain('www.fireflake.com');
?>
</div>

Sure, it will still show up in the code but after this code is run you will have the response time in the $status variable and the possible error message will only be visible to someone who inspects the code! If $status is -1 the server is unreachable and 0 and up is the response time.

Also, due to the fact that DNS servers usually return a “search engine” response when a domain is unknown, you are better of using IP-numbers for checking that servers are truly on-line or not (I hope they have a fixed IP!). If not you would have to read a file from the server and make sure it has the contents you expect to be sure you are on the right server and your DNS isn’t fooling you.

Not exactly software but technical at least so I just have to link it.

So if we do manage to make a black hole and suck us in this is the last remnant of our civilization. It’s a valid contribution.

More and more I’ve come across situations where I want to use AJAX to download a XML file to use in the interface but know beforhand that the file will take a long time to load. With asyncroneous download of XML files by JavaScript, which is kind of what the buzz word AJAX is all about, you must be carefull not to leave the client in limbo between a useable interface and a locked up screen.

Unfortunately this script only works in Internet Explorer, useful tips of how to port it properly (with the asynchronous property intact) would be highly appreciated.

Here is a simple description of the basic functions needed to perfom a asynchronous download where the user will have the option to abort.

First we need a simple function that download the XML, this is pretty standard and the code is lovingly ripped off from w3school.com.

function loadXml(sUrl){
	try{
		//Internet Explorer
		xmlDoc=new ActiveXObject("Microsoft.XMLDOM");
	}
	catch(e){
		try{
			//Firefox, Mozilla, Opera, etc.
			xmlDoc = document.implementation.createDocument("","",null);
		}
		catch(e) {
			alert(e.message)
		}
	}
	try{
		xmlDoc.async = 'true';
		xmlDoc.load(sUrl);
	}
	catch(e) {
		alert(e.message)
	}
}

This code is pretty straight forward and I assume you allready know of it, if not read the guide over at W3Schools. The only difference in the above code compared to that from the tutorial over at W3Schools is the flag “xmlDoc.async = ‘true’”. This means that the code will continue executing after the load is called without waiting for the load to finish. This will place the status of the xmlDoc variable in a limbo which can be checked with the “readyState” flag.

To check if our file is ready to use we have a test-loop that will check if readyState changes:

function testReadyLoop(){
	i++;
	if (xmlDoc.readyState == 4){
		// the file has completed the download
		alert('xmlDoc ready to use! Contents:\n' + xmlDoc.xml);
		// TODO: add code here of what to do with the file
	}
	else{
		if (!abortXmlLoad){
			// try again in 1 second
			setTimeout("testReadyLoop();",1000);
		}
		else{
			// stop loading the xml file
			xmlDoc.abort();
			alert('Loading of the XML file aborted!');
		}
	}
}

The incrementation of the variable “i” is just a counter that will be used later and the “abortXmlLoad” is a boolean if the loop should continue or not, these will be explained later. What happens in this function is that it first tests if readyState is 4 which indicates that the file is ready to be used, if this is the case we simple show an alert with the contents of the file, here more intelligent code would be placed. If it’s not ready it checks if it should continue waiting for the file or not, if it should it calls itself in 1 second (1000 ms) otherwise it aborts the loading and simply stops.

To abort a download we need to set the “abortXmlLoad” flag to true, a short function is needed for this:

function abortAsyncXML(){
	// set the abort flag to true
	abortXmlLoad = true;
}

Now we have all the functions needed for the asynchronous download, a last function is added to tie them all togheter:

function loadAsyncXML(sUrl){
	// set abort to false and start download
	abortXmlLoad = false;
	i = 0;
	loadXml(sUrl);
	// start loop to check when ready
	testReadyLoop();
}

This function first resets the values of “i” and “abortXmlLoad” and then it calls the download and after that starts the loop to test if the download is ready. The file will now download silently in the background and pop an alert when ready unless someone calls “abortAsyncXml” before that happens.

As you may have noticed there are a few global variables I use across the functions that also need to be added to the top of the script:

var xmlDoc;
var abortXmlLoad;
var i;

To use this script a small form need to be added to the page:

<form>
<input type="button" value="load" onclick="loadAsyncXML('sample.xml');">
<input type="button" value="abort" onclick="abortAsyncXML();">
</form>

This will load the file “sample.xml” and abort if the abort button is pushed. In order to test that the abort button is working you would have to build a slow loading page that simulates long loading time.

I will post a link to the full code and sample later. Hope you found this helpful.

Are software patents protecting the inventors and is something needed to protect intelectual property?

Read this: http://news.zdnet.com/2424-9595_22-218626.html

If it designed to be used to protect intelectual property then it is now abused by large companies for making absurd claims, in this case Microsoft.

Makes me want to patent the “O”-button… peple cant write blgs withut it.