Performance optimizations aside, and even if you concatenate all your JS files into a single file, you still have to load the library into the browser after the request has been made, and you can’t do anything with it until that has finished loading.

Wouldn’t it be so great if the browser had a copy of YUI or jQuery or Dojo or Prototype or GWT in memory at all times? Then your library using decision would not be influenced by the size of the library but by its speed and performance, and perhaps features and functionality provided. I mean you don’t go loading new fonts and new plugins every single time you go to a new page right? Then why do we have to download the same 5 or 6 JS libraries over and over all day long as we browse the web? And if this was an option how many companies would spend time rolling out their own rather than adopting one?

I understand there would be tons of politics involved and it will probably never happen in a million years but it is worth a try. Perhaps at least a FF only version could be done through an add-on that loads the latest major version of all these libraries and checks for updates on startup, similar to how other plugins get updated in FireFox today. The browser would have to namespace the libraries to avoid confusion. Developers would then just do a test to see if the browser has a required library in memory, and if not then make a server call for it as a fail safe.

It would take lots of collaboration and standardization. JavaScript was the sleeping giant that was brought to center stage with the success of frameworks and libraries that have fueled its standardization and resulting wider adoption. Why not give those libraries some credit and package them into your browsers?

It turns out, it is really simple to do. The trick is using the IMAP feature of Gmail. Here is how to import your emails from your current email client to Gmail. These steps assume you are using Outlook, but they should be pretty much the same for any other email client including Thunderbird, Entourage, Mac Mail etc. So let’s just get to the steps.

1. Turn on IMAP in your Gmail account: In your Gmail account go to Settinngs > Forwarding and POP/IMAP > IMAP Access > Enable IMAP > Save Changes
2. Using the settings for Gmail IMAP for your Email Client, connect to your Gmail account using a new IMAP account. Once you’ve this done, you will see your Gmail account with labels listed as folders in your email client.
3. Now just highlight the emails you want to migrate to Gmail in your Personal folders and drag and drop them from your local account to the Gmail IMAP folders.

That’s it! The emails get moved to Gmail with all the header in tact. It might take some time for the action to take place, if you have tons of email to migrate.

Notes:
You can drag them to any folder (label you want) and you can also create new folders as you wish, which will become labels in Gmail. If you don’t want your old email to clutter your inbox, you can just drag them to the Gmail/All Mail folder which basically archives them and removes the Inbox label.

Another cool feature in Gmail is support for POP3 downloads from other accounts and additional SMTP accounts for sending emails. These can be configured under Settings > Accounts and can be good if you want to combine and consolidate other existing emails into your Gmail account. If you want to go all the way there is even a labs feature called Multiple Inboxes that can add a new column with configurable filters to the right side of your Gmail Inbox.

The night started out with Finnish guys Rotten Sound which put on a solid performance. Next Aborted picked up and man those fellas really sounded good live, one heck of a pit frenzy. Chatted with some old school and new school fans in and around the show during the breaks. Saw Shane from Napalm hanging out at a T-shirt booth with a sign behind him saying “Don’t Ask Stupid Questions” which was pretty hilarious.

After Aborted 1349 went on which I had never seen live before. They use face makeup and are pretty fast for a black metal band. Suffocation presented a nice change in the pace of the pit and I have to say that was the only band that I actually understood what the singer was saying when talking to the audience between songs. They really grinded up the crowd when the played Infecting the Crypts.

Then finally the moment of truth, after a 30 minute setup period. CAR CASS CAR CASS chants started. There was a very creative intro with a formal sounding lady’s voice with a British accent reading a very morbid script. Really cool stuff, I wish I had that script or hear it again. Then for the first time in more than 15 years Carcass came out and they played one heck of show. Last time I had seen them was with Death (RIP) and Pestilence back on 10/05/1990 at the legendary Country Club (used in the Boogie Nights movie) in Reseda, which is no more. I can’t say enough about how truly good Carcass was last night. Truly a legendary band with a classy performance. Bill Steer, Michael Amott, Jeffery Walker, and Daniel Erlandsson (deep resume) subbing in for Ken Owen who is out due to a brain hemorrhage. My prayers and best wishes go out to Ken. They played lots of their classics like “Incarnated Solvent Abuse” and “Exhume To Consume” and amazingly done. I would say the sounded even better than they did last time I saw them live.

So the lucky part was at the end when Daniel came out and threw out his drum sticks and one landed almost perfectly vertically and between my two feet. This was with about 50 other people trying to get it with their hands up in the air. I was so happy!! Here’s a photo of it, I will cherish it for ever.
* WPG2 Plugin Not Validated *

Long live the English Death Grind core Kings Carcass. I’m predicting this tour will somehow influence them to create a new album once they get back home. They are playing tonight at the Grand Ballroom in San Francisco. Don’t miss it! It might be the last. Here are a few shots I took last night with my phone.
* WPG2 Plugin Not Validated ** WPG2 Plugin Not Validated ** WPG2 Plugin Not Validated ** WPG2 Plugin Not Validated *

More old school death metal is on the way, Obituary are set to play HOB next Wednesday 9/24/08. I will try to make it there.

As if it is not bad enough that they have a pretty annoying and unusable multi-step, multi-page ticket sales process, recently they are starting to give out some borderline non-sense captcha validation images. How many people would just say “forget it” and leave? Well not many, since TM is an undocumented Monopoly for ticket sales in my opinion. Pretty bad user experience overall. I would love if they at least put the reCaptcha page on the same page as the ticket search form to cut down page loads. Better yet, make the whole process an ajax based system so you never have to leave the order page. It sucks to go through a blinding reCaptcha page and then get the dreaded “You’re Screwed” all sold out page. Just tell me right upfront: “Hey Don’t bother, no seats are available mr!” and don’t waste my time. I mean I have to pay an inflated ticket price + order processing fees + convenience fees + delivery fees. Can we at least get a convenient to use website?

I went ahead and did a 2 minute reshuffling of the order form for TicketMaster and this is what I came up with. Just put all forms on one page and got rid of the clutter. Progress bar and search results would appear on the same page.

I noticed also that there are now TicketMaster hosted auctions and scalping services on site. Wonder how far they have to push the envelope before regulators are forced to step in.

Here is an alternate pwned ending version, pretty weird.

Brand new Ad

What we knew so far about Chrome

• Uses V8 JavaScript engine which supports Classes and compilation. There is a V8 JavaScript benchmark suite that gives FireFox 3 a score of 83, while giving Chrome a score of 1213! Safari 3 gets a score of 128. Judging by this alone, V8 blows away the competition.
• Uses Webkit rendering engine.
• Tabs run as independent processes which can be managed.
• Lots of other features.

First Impressions

• Installation was a breeze and imported FireFox settings.
• No status bar, You only see the status bar when you hover over a hyperlink.
• The Task Manager (Shift + Escape) updates in real time and shows memory, CPU and network usage for each tab, each plugin, and the main Chrome process separately. There is also a link to Stats for Nerds with lots more gritty info.
• The Flash plugin is extremely CPU usage intensive and causes sluggishness when scrolling. I just loaded a popular flash website and noticed my machine came down to a near halt. It seems to happen more with Flash files that contain infinite loops, using as much as 70% of the CPU.
  
• The built in JavaScript console looks like a combination of FireBug and Web Inspector.
• There is a built in JavaScript Debugger (Alt + `)
• Passes the Acid 2 test.
• Chrome gets a score of 78 on the Acid 3 test, which is higher than FireFox 3 at 57, Safari at 72, and Opera at 45.
  
• Omnibar - this is the URL/location bar in Chrome that has some fuzzy logic built in to suggest “smart” autocompletes. This is the current order of the drop down in the auto complete list. There seems to be no way of changing this ordering as of now. Would be nice to be able to customize them.
   
  
   
  1. Search Google for FOOBAR
  2. FOOBAR/ (I’m not sure how useful this one is really)
  3. Link to the FOOBAR Wikipedia page
  4. Link to I’m Feeling Lucky URL for term FOOBAR. This item gives you the ability to search within the URL. For example if you type in Amazon in the Omnibar and select the amazon.com option using the down arrows, you will see “Press Tab to Search Amazon”. See below instruction on implementing this search functionality for your website.
  5. Search Google for FOOBAR ANOTHER TERM
  6. Search Google for FOOBAR ANOTHER TERM
  7. A page in your history pertaining to FOOBAR
  8. Link to history search for pages about FOOBAR

Cool Developer related stuff in Chrome

Google Chrome User Agent String:

Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/525.13 (KHTML, like Gecko) Chrome/0.X.Y.Z Safari/525.13.

Chrome has a menu option called “Create application shortcuts…” that uses Google Gears to create a shortcut to your webapp. Users can choose to place the shortcut to your webapp on their Desktop, on the Start menu and even the Quick launch bar in Windows. This is a pretty powerful feature. When a user clicks this icon, Chrome opens up without the Omnibar, and your website will appear in an “Application” format. You can customize how Chrome creates these shortcuts using meta tags. These tags are named: application-name, description, application-url, and shortcut icons in both 32×32 or 48×48 formats. The favicon is used if not specified. For example you can use the following HTML code in the head of your document. Note that with the Mozilla added support which is used in Chrome, you can use any supported graphic format as your favicon, and not just the old school favicon.ico file.

<head>
  <meta name="application-name" content="Gmail"/>
  <meta name="description" content="Google's approach to email"/>
  <meta name="application-url" content="http://www.gmail.com"/>
  <link rel="icon" href=gmail_32x32.png sizes="32x32"/>
  <link rel="icon" href=gmail_48x48.png sizes="48x48"/>
</head>

To open a new tab from your webapp in a separate process using JavaScript you can do this in Chrome.

var w = window.open();
w.opener = null;
w.document.location = "http://differentsite.com/index.html";

Chrome lets users search your website from its ominbar. To enable and include your website’s search in Chrome you have to create an OpenSearch description document (OSDD).
For example you can create something like this:

<?xml version="1.0" encoding="UTF-8"?>
 <OpenSearchDescription xmlns="http://a9.com/-/spec/opensearch/1.1/">
   <ShortName>Web Search</ShortName>
   <Description>Use Example.com to search the Web.</Description>
   <Tags>example web</Tags>
   <Contact>admin@example.com</Contact>
   <Url type="application/atom+xml"
        template="http://example.com/?q={searchTerms}&amp;pw={startPage?}&amp;format=atom"/>
   <Url type="application/rss+xml"
        template="http://example.com/?q={searchTerms}&amp;pw={startPage?}&amp;format=rss"/>
   <Url type="text/html" 
        template="http://example.com/?q={searchTerms}&amp;pw={startPage?}"/>
   <LongName>Example.com Web Search</LongName>
   <Image height="64" width="64" type="image/png">http://example.com/websearch.png</Image>
   <Image height="16" width="16" type="image/vnd.microsoft.icon">http://example.com/websearch.ico</Image>
   <Query role="example" searchTerms="cat" />
   <Developer>Example.com Development Team</Developer>
   <Attribution>
     Search data Copyright 2005, Example.com, Inc., All Rights Reserved
   </Attribution>
   <SyndicationRight>open</SyndicationRight>
   <AdultContent>false</AdultContent>
   <Language>en-us</Language>
   <OutputEncoding>UTF-8</OutputEncoding>
   <InputEncoding>UTF-8</InputEncoding>
 </OpenSearchDescription>

Fore more tips read the Google Chrome FAQ for web developers page.

ps. Don’t use the Google gears Chrome download page which gives a JavaScript error!

_GU_SetupOneClick is not defined
onload(load )

arguments itself returns an object that looks like an array (but not really an array) of the arguments passed to the function.

Prior to JavaScript 1.4 the Function object also had a similar arguments property, which is now deprecated.

However the Function object comes with a few other useful properties that we can still use to get argument related data.

function callTaker(a,b,c,d,e){
  // arguments properties
  console.log(arguments);
  console.log(arguments.length);
  console.log(arguments.callee);
  console.log(arguments[1]);
  // Function properties
 console.log(callTaker.length);
  console.log(callTaker.caller);
  console.log(arguments.callee.caller);
  console.log(arguments.callee.caller.caller);
  console.log(callTaker.name);
  console.log(callTaker.constructor);
}
 
function callMaker(){
  callTaker("foo","bar",this,document);
}
 
function init(){
  callMaker();
}

For demonstration purposes, you can run the init function above and view the logs in FireBug.

arguments object and its properties

arguments returns ["foo", "bar", Window, Document]

arguments.length returns 4

Note: even though our function has a signature with 5 arguments, length returns only 4 here. This is because the caller sent us only 4 arguments. See below for how we can use Function’s length property to find the number of expected arguments.

arguments.callee returns callTaker(a, b, c, d, e)

Note: callee shows us the signature of the currently executing function and is useful when trying to make recursive calls to a function within its own body.

arguments[1] returns bar

Note: arguments can also be set for functions in an array like format. For example you can set the second argument like this: arguments[1] = ‘moo’;

Function object and its argument related properties

callTaker.length returns 5

Note: This is the expected number of arguments.

callTaker.caller is the same as arguments.callee.caller and returns callMaker()

Note: we can go up the stack trace and get the caller of the caller etc. For example we can find the function that called callMaker using arguments.callee.caller.caller which returns init().

callTaker.name returns callTaker

callTaker.constructor returns Function()

Note: Since we have not modified the basic behavior, we see the built in function that creates an object’s prototype for our function, which is the Function object.

Basic Usage Sample

var dataArray = ["One", "Two", "Three", "Four"];
 
var lister = function createList(list) {
 if(arguments.length == 3){
  var result = "<" + arguments[1] + ">";
  for (var i = 0; i < arguments[2].length; i++){
   result += "<li>" + arguments[2][i] + "</li>";
  }
  result += "</" + arguments[1] + "l>";
  document.getElementById(arguments[0]).innerHTML = result;
 }
}
 
function makeList(){
 lister("list_HTML","ul",dataArray);
}

Run the sample

References:
JavaScript Stack Trace
Function
arguments

Take a look at some of the older user interfaces to get a sense for how del.icio.us has evolved until now. Aside from the new look there are new features as well. The delicious dev team has supposedly completely rewritten the code base, to allow for a more scalable and spry tool.

New features:

Navigation: New simple CSS tabs with onclick drop down menus make is simple to find your tasty bookmarks. The navigation structures seems to borrow from Flickr. Search also comes with a simple drop down option list to find stuff in your bookmarks, network or everyone.

Bookmarks: You have Title view, regular view and Full View options here. View your popular bookmarks to filter the good stuff and filter down more by choosing fresh only.

Sidebar: is updated to show a relative set of tags related to your left content pane.

Action Box: This light blue box in the sidebar shows the common tasks for the given page.

Search URL: This is a reverse search for bookmarks and pulls in the ones with the given URL along with tags used for that URL by each user, broken down in chronological order. You can also view the notes added by each user to the given URL.

Settings: The settings page is laid out with all the actions listed on one page, similar to a sitemap.

Forums: There is a support forum that looks more like a categorized list of links than a forum. I like the non-cluttered design. Reminds me of Drupal.

As in the past their API is still available for developer consumption.

margin-top | margin-bottom: max( marginbox 1 … marginbox n)

When it comes to the real world there are 3 major categories of margin collapsing. So let’s figure out how these 3 types of collapsing work and how to tame them.

1) Adjoining boxes of sibling elements:

The bottom margin of an in-flow block-level element is always adjoining to the top margin of its next in-flow block-level sibling, unless that sibling has clearance.

 <div>
  <div>content</div>
  <div>content</div>
 </div>

While margin collapsing is great for written text such as paragraphs and headings etc., it can get somewhat tricky if you’re trying to get pixel perfect spacing between your boxes, so there might come a time when you want to disable margin collapsing. Take a look at these examples that suggest ways of uncollapsing your margins.

So it seems the only way to break out of collapsing that works in both IE and FF is our second example by using floats and clear combined. You might have noticed the above 4 boxes themselves also have 20px margins which have collapsed to form 20px margins within our 3 adjoining sections in the middle.

2) Adjoining boxes of ancestor elements:

The top or bottom margins of contained elements will always collapse together to form one margin. If the element’s margins are collapsed with its parent’s top margin, the top border edge of the box is defined to be the same as the parent’s. Below is the sample HTML and diagram and explanation of how to avoid this.

 <div>
  <div>content</div>
 </div>

Remember that collapsing only happens if the margins actually touch one another. In this case the inner box’s margin ends where the outer box’s margin starts, therefore they are touching. One easy way to avoid this type of margin collapsing is to add either vertical borders or vertical padding to the parent box. Take a look at these examples:

3) Self collapsing boxes:

Empty block level elements result in a special case of margin collapsing. Their top and bottom margins actually touch each other and they basically self destruct based on the rules of margin collapsing. According to the W3C: An element’s own margins are adjoining if the ‘min-height’ property is zero, and it has neither top or bottom borders nor top or bottom padding, and it has a ‘height’ of either 0 or ‘auto’, and it does not contain a line box, and all of its in-flow children’s margins (if any) are adjoining.

 <div>
  <div>content</div>
  <div></div>
  <div>content</div>
 </div>

Another obvious way to break out of collapsing is changing the display to absolute. Finally, note that the margins of the root element box in any document never collapse. There are a few rules for negative margins when it comes to collapsing. I will try to write another post on this later.

I’ve created a demo page here with the above examples.

References and Acknowledgments

W3C Box Model Collapsing Margins,
Andy Budd No margin for error,
Research Kitchen CSS Autoheight and Margin Collapsing,
Complex Spiral Uncollapsing-margins,
La Chatte Noire tests: block formatting context, display:inline-block and margin collapsing