The Savage Doctor

Discovering XML and XSLT

2005-07-01T12:30:00.000+08:00

As a web developer, I have to learn about new technologies that may (or may not) be applied in my job. Along my journey of self-teaching discovery, I've learned PHP/MySQL, ASP, new-fangled JavaScripts, standards-compliant HTML, Web Accessibility Guidelines, etc.

I've also tinkered with XML, e.g. for managing information where databases were impractical. That's really all I saw XML as: a database replacement. On a less conscious level, I knew that it could be put to better use, e.g. HTML is a derivative of XML and web pages (which use HTML) are not databases.

Then yesterday, I took another look at something that I had only given a cursory glance previously: XSLT. XSLT is to XML what CSS is to HTML. And if all that is gobbledy-gook, suffice to say that XSLT/CSS are how a page looks-and-feels, while XML/HTML is the actual page content (i.e. separating form (XSLT/CSS) from function (XML/HTML)).

I've since learned that XSLT, unlike CSS, can also include HTML/XML(?) content, and this will display properly in a (XSLT-supported) web browser! Look at this XML+XSLT page. (Yes, it's hosted on my company's server and uses the corporate website's look-and-feel, but it's just a test, so it should be fine (and I'll probably create a more permanent test page somewhere else when I have a few minutes to spare).)

(The page was tested with Firefox and Internet Explorer 6 SP 2. It should also work in Safari 1.2 and other modern web browsers.)

"What's the big deal?" you think. It looks like an ordinary web page, it acts like an ordinary web page. The big deal becomes apparent when you look at the page source (Firefox: right-click in page, select "View Page Source"; Internet Explorer: right-click in page, select "View Source"). Seems like an awful small chunk of code, right?

And that's the beauty of XSLT! The heavy lifting, i.e. displaying the page elements, including common images, links and columns, is done in XSLT. I could create another XML file with just as little (or more) code, link it to the same XSLT file, and you will see the same layout with whatever content is in that other XML file.

Again, you may ask, "So what?" Well, most web pages on the same site usually share some common elements, e.g. logos, menus, footers, etc. In traditional HTML, these common elements have to be coded into each and every one of the pages. There are workarounds if you use dynamic scripting (PHP, ASP) or you might even get away with using JavaScript.

If that is too technical, then think of web pages as dishes of food, e.g. chicken, beef, fish, pork, etc. garnished with the same vegetables. What you would do is prepare all of the vegetables, distribute them among the different dishes, then cook the meats separately and scoop them into their individual dishes. Now all of your dishes will have the same vegetables but different meats, and it would've taken a much shorter time to prepare than if you had cooked the vegetables for each dish one-by-one.

If my example is clear, then you should be able to see that the vegetables are XSLT and the meats are XML. You design a XSLT file, then create a bunch of different XML files, then link the XML files to the XSLT file. Voila!

And because XSLT is based on XML, it can be used with any (XML-supporting) tools. Sweet! Of course, I could create HTML-based templates to apply to all of my web pages. However, those templates would only be recognisable by Dreamweaver. Which is not good if I decide to ditch Dreamweaver.

One more bonus: by stripping the common code from each web page, XSLT reduces the "page weight", i.e. the size of each page (and therefore the loading time). (The XSLT file is cached by the web browser after its initial loading, so it doesn't have to be transferred over the Internet every time.)

Of course, XML and XSLT (and other XML-related goodies) can be used for other purposes, not just displaying web pages, but this is a step-by-step journey. Also, I haven't mastered everything about XSLT, e.g. how to set images and links in the XML file (but they're described in this IBM XSLT page). But I believe that I'm heading in the right direction.

Onward, ho!

--

Technorati tags: XML, XSLT

Web accessibility standards now!

2005-02-17T18:58:00.000+08:00

When you think of browsing a website, you think about sitting in front of your computer, starting Firefox (or Internet Explorer, for those of you still living in the Dark Ages), and going to a website, like Yahoo!, and admiring its pretty little pictures. Right?

Wrong. Blind people have web browsers that don't show any pictures. People who don't have 20/20 vision may increase their browsers' default text size to something larger than what you and I would use. And then, in the 21st century, there are web browsers in personal digital assistants, handphones, what-have-yous, with their little screens.

And that's not to mention the myriad of web browsers with different interpretations of something as simple as how a bulleted list should display.

So the big question for web designers/developers is: Do I design for

a particular web browser, like Internet Explorer, or
as many users/uses as possible?

Designers/developers who don't know better, who unfortunately make up more than half of all designers/developers, choose (a). And then there's the rest of us... who unfortunately design for web browsers for computers and not thinking about PDAs, smartphones, screen readers for the blind, etc.

That's why the World Wide Web Consortium (W3C) came up with its industry standards. Based on my current understanding, there are two standards that any and all web designers/developers need to be concerned about:

XHTML 1.0 - which specifies the language behind the web page, and more importantly,
the Web Accessibility Initiative - which specifies that web pages are as accessible as possible by any living human being, regardless of their bodily health and device used to access those pages.

Fortunately, it's easy to check that a web page is XHTML-compliant. You just run a web page through a XHTML validator and it'll tell you what syntax mistakes need to be corrected. Simple.

But to check for WAI compliance, more work is needed. There are some syntax mistakes that can be caught automatically, like accesskey and tabindex. But how do you deal with things like colour contrast? Currently, there are no regular machines that can take an image, analyse it, and tell you how the contrast fares. (If there are such regular machines (not supercomputers!), please let me know.) The only way widely available at the moment is the human eye.

And then there are other things. Even things as stupidly simple as... text sizes. Web designers/developers work on a known element: their screens. But their web pages are displayed on an unknown element: other people's screens. You simply cannot control what other people use to look at a web page. Not even if you're Bill Gates.

As a result, text that look "just right" on one screen look too big or too small on another. The solution is to use relative text sizes, e.g. in percentages. But this can still mess things up. For instance, if you have a box of text at the top right corner of your page, it may take up a small amount of space or it can be so large that it takes over half the screen. Why? Because different people specify in their web browsers different text sizes that they are comfortable with, so the box "grows" to accomodate this. So what looks good on paper looks horribly, horribly wrong on screen.

What's the ultimate solution? As long as we want "pixel perfect" web pages, there will be no ultimate solution. Instead, we must adapt to the fluid nature of the web. And for web designers/developers, it means making some hard decisions, like designing flexible page layouts. It's a pain, but a necessary one, if we want to ensure that web pages look good today on the majority of browsers, and look just as well in the future.

Web designers/developers must embrace the WAI's guidelines (in addition to the XHTML guidelines) today.

--

Technorati tags: web design, web development, web standards, webmasters, accessibility, W3C, XHTML, WAI

The case for the iPod shuffle

2005-01-15T16:46:00.000+08:00

By now, most people in the developed world should know about the iPod shuffle. And there have been some complaints about it. From what I've read, the most common complaints are:

There is no screen.
It has limited storage space.
It randomly shuffles its songs.
No one wants to wear it.
It isn't well-designed.

All of these are valid complaints in their own right. Compared to the iPod and iPod mini, the iPod shuffle is definitely crippled. But are they justified? Is Apple pushing a product that has no market?

I think not. Almost all of these complaints have come from one very vocal group: men who are technologically-inclined. But as a Singaporean credit card advertisement goes, "the men don't get it." I believe that Apple's target market isn't just people who want a portable digital music player.

No, I belive that the target market comprises:

people who have small song libraries
people with a predetermined playlist of favourite songs
people who already recognise and know their songs
athletes or people who are working out or constantly on the move
people at parties who become impromptu DJs
young, fashionable women.

That last line is key. Often times, electronics are targeted at one group: men who are technologically-inclined. These are people who not only understand electronics instantly, but are willing to pay more for the fastest, largest, flashiest, best electronics. But it is a common marketing and economical fact that men don't spend as much as women.

Women, especially young, trendy ones, are the big spenders, not only in fashion, but in food and entertainment. They are an untapped market by many big electronics makers... and they have immense spending potential.

With these target market requirements in mind, here are my arguments against the complaints stated above:

1. "There is no screen."

The iPod shuffle doesn't need a screen because its owners already know what songs are in it and don't need to be reminded. All a screen will do is steal precious battery power that could be used to play more songs.

When you're playing your favourite songs or a small library of songs, you are already familiar with those songs and will hardly ever need to look at the screen, for example, to see the title or artist, because you already know these details. In fact, your player will probably be sitting comfortably in a pocket or bag, and fished out only when the volume needs to be adjusted or you're done listening to it.

For a limited number of songs, you don't need a screen.

2. "It has limited storage space."

Again, this goes back to the small number of songs that you will listen to. Don't laugh, but some people have small song libraries. Or they have large libraries, but only a few favourites that they listen to repeatedly.

The iPod shuffle's smallest capacity is 512MB. Apple estimates that you could fit 120 songs in it, although some real world tests have found that you can fit more, like Walt Mossberg's reported claim of 188. But even with 120 songs, at four minutes each, that's 480 minutes, or eight hours.

That may not sound like a lot, until you think about the daily commute to and from work. Suppose you spend one hour going to work and one hour returning home each day. That's two hours of commuting time a day. With the iPod shuffle, you would have enough music to listen for four days before needing to fill it with new songs. And four days is just one day short of an average work week.

Remember also that the iPod shuffle's battery life is rated at 12 hours. So you can repeat half of the songs after the first eight hours and, chances are, not be overly bored because these will be some of your favourite songs.

Yes, the amount of space is small. But it's sufficient for the few favourites that you can listen to on the bus or train.

3. "It randomly shuffles its songs."

Given that you will be listening to a subset of your song library (or your whole library, if you have a small one), you will get bored of listening to the songs very quickly if they are played in the same order each... and... every... single... time. Man needs some variety in life to keep him going.

Listening to the songs in the same play order can lead to song fatigue, which may even lead to your disliking what used to be your favourite songs. Shuffling the playing order helps to keep your song playback interesting.

If you have 120 songs in the 512MB iPod shuffle, the playback order could be different thousands of times. (My math isn't as good as it used to be, but I think the number of different combinations could be 2 to the power of the number of songs, i.e. 120. Which is a huge number.) And this will introduce you to different ways of ordering the playback of your songs.

For instance, all along, you may have had this playback order: A-B-C. After shuffling, the order is C-A-B. And you may then realise that song A sounds pretty good after song C. And then you shuffle again, getting B-A-C, and realise that song B makes for a pretty good opening song, like the opening theme of a good movie.

Of course, the reverse could happen, i.e. the random playback order may mismatch songs, like a concert piece after a rap. But the variety that randomness brings introduces you to new song mixes. And you learn to better appreciate your songs.

Life isn't just random. Life needs change to be engaging.

4. "No one wants to wear it."

This line should really read: "No man wants to wear it." And this would only be true for a subset of men, mainly, those who are middle-aged or older, though seniors may like to wear it so that they don't lose it, the way they hang their spectacles around their necks.

But young women like anything that is a fashion accessory because it draws attention to them. I'm not saying that all young women are vain or fashion-conscious, but in a consumer, materialistic world that emphasises beauty and attractiveness, anything that enhances those two qualities is deemed as desirable.

Young, trendy women will wear or display anything, from jewellery to handbags to handphones. Take the handphone. Women will rarely buy a handphone because of its feature set. They will first judge its design and whether it complements them. If it does, then they will look at the feature set. Many times, women will pass on a powerful phone because it's too bulky or it lacks that certain look... or even because it has a plastic case.

Nokia itself is getting into this electronics-as-fashion-accessory craze. Its line of "Distinctly Bold" phones, with their black-and-red stylized designs, are designed to appeal to fashion-conscious people. Internally, they are still regular handphones. But it is their external design that sets them apart from others, and makes them more engaging.

Similarly, the iPod shuffle is an accessory for these young women. Because it has a striking white colour and can be worn around the neck with a lanyard, it becomes something that makes them stand out from the crowd. Of course, there is the secondary advantage that you are less likely to lose something that small when you wear it around your neck.

People, especially women, will want to wear it, not just because it's convenient, but because it's eye-catching.

5. "It isn't well-designed."

Firstly, design is subjective. What looks good to one person will look like crap to another. For every supporter, there is a detractor. No single design can win over everyone. So it is for the iPod shuffle.

Having said that, there are a few points to note about the iPod shuffle. It has a non-bulky, thick pen-like design, which reduce its volume and weight. This makes it more convenient, for instance, to be worn around the neck or kept in a purse. Yes, it resembles a tampon or pregnancy kit. Is it no wonder then that I think it is targeted at young women who already purchase such items, and would appreciate a music player that is equally non-intrusive?

Also, given its small size, it can be used easily by athletes or people who are exercising or constantly on the move. Especially when working out, you don't want anything that weighs you down and forces you to exert more effort unnecessarily.

Small size need not be a negative characteristic. And a long-ish, thick-ish finish need not be a bad design.

So, given what I believe is Apple's target market for the iPod shuffle and the iPod shuffle's own characteristics, I don't agree fully with the complaints against it. I believe that there is a market for the iPod shuffle. It's just not so obvious at the beginning.

--

Technorati tags: iPod shuffle, music, fashion

Why M1 may win the 3G war in Singapore

2004-11-24T08:24:00.000+08:00

(This post was moved from my LiveJournal.)

I'm going to go out on a limb here and say that M1, Singapore's second largest mobile operator, will "win" the 3G war.

A bit of background: The local telecoms authority, Infocomm Development Authority (BTW I still scratch my head over why Singapore must be so special and invent the word "infocomm"), has directed all of our mobile operators to get their 3G services up and running by the start of 2005. About time!

Since 3G is a mobile service, every operator around the world has focussed on, of course, handphones. But 3G does more than allow you to make phone calls; it can also zip data around wirelessly at up to 384kbps (under optimal conditions). By comparison, most DSL connections start at 256kbps.

(BTW by going online through 3G, I don't mean through handphones, but through PC cards inserted into laptops or with PDAs that have 3G capability.)

What does this mean to the mobile computer user? Fast data transfers anywhere and everywhere! But wait! What about those wireless hotspots? These use industry standard 802.11 technology, with data speeds starting at 11Mbps, or about 11,000kbps. That's slightly under 30 times the top speed that 3G promises. So why do I think 3G has a snowball's chance in hell of beating these hotspots in luring road warriors away?

Well, think about it. Whip out your handphone anywhere in Singapore and you have a 99 out of 100 chance of making a call to anyone, anywhere. Since 3G is a mobile service (and if the operators get their acts together), you will likewise be able to whip out your gee-whiz 3G handphone anywhere in Singapore and make a phone call with little to no problems too. Now if you're using 3G to transfer data wirelessly, then you can likewise be able to go online wherever you can make a 3G phone call.

Too many words? Algebra might help.
Let X = 3G is a mobile service
Let Y = phone calls through 3G can be used anywhere problem-free (under optimal conditions)
Let Z = 3G can transfer data at up to 384kbps, or at a pretty fast speed
X is true because it is an industry standard. So if Y is true, then Z must also be true. (In Singapore, it is difficult for Y to be false due to the IDA's 3G guidelines.)

In constrast, hotspots are provided by service providers and limited to a certain area. Instantly, you can see two problems with hotsposts:

You need to find a hotspot served by your service provider.
If you leave the hotspot, you've lost your online connection.

I think these two problems are the main reasons why wireless hotspots never really took off in a big way, and provide a huge opportunity for 3G.

M1 has already announced trials for its 3G data service. This is unlike what the other two operators are doing, mainly focussing on handphones and how people will want to either videoconference while on the move (can people walk, talk and look at two different places at the same time?) or be entertained 24/7 (on small screens and tinny speakers? I think not!). In light of my argument above, I think data service might prove to be a more lucrative 3G-based income generator.

So does this mean that I'll be first in line to sign up for 3G? Hell no! Product and service providers are always talking about pricing their goods or services at a price that the consumer is willing to pay for a reasonable return to the providers. I think that consumers are willing to pay a certain cost based on how much disposable income they have available. Since I'm not filthy rich, and since 3G is likely to be prohibitively expensive initially, I don't plan to get 3G in spite of the promise of a mobile online connection.

So there. My (virtual) money's on M1 to succeed at 3G.

--

Technorati tags: 3g, Singapore, M1, wireless Internet

What is The Savage Doctor?

2004-11-01T01:00:00.000+08:00

From Dictionary.com
sav·age: n. A person regarded as primitive or uncivilized.
doc·tor: n. A person, especially a physician, dentist, or veterinarian, trained in the healing arts and licensed to practice.

It has been said that we live in interesting times. Twenty-first century technology is such that it changes at such a rapid pace that it becomes difficult to keep up with what is useful and what isn't. We are overwhelmed by information and appliances. We are like cavemen discovering the light bulb.

Yet we need to make informed decisions so that we can live the best lives that we can. We cannot afford to go on without knowing what information and technologies will have a significant impact on our lives. Using the wrong technology would be akin to inflicting a terrible disease on ourselves.

We are savages who have succumbed to the illness of information overload and require a doctor to become better.

We need The Savage Doctor.

And that is what this blog aims to be. Granted, it is a lofty goal for someone who is not necessarily involved intimately with the technological world. But like any good doctor, I read and listen and evaluate, and finally come to a decision that I believe will improve the value of a person's life, particularly my own.

This blog isn't a fanciful review site. It also isn't the blog of a technological elite. Neither is it a blog where you can find the latest and greatest news, let alone one that is published every day. It is a blog of a person who wants to make sense of information and technology to decide how it can improve my life.

Welcome to my blog.

---

On another note, the inspiration for this blog's title comes from a famous DC Comics villain, Vandal Savage. Savage is a Neanderthal who was granted immortality by an extraterrestrial meteor. Since then, he has manipulated leaders and nations to achieve his aim of world domination, but is always foiled by heroes, super or otherwise.

I took the name "Savage" because the villain's drive inspires me to reach greatness, though not as a super villain. Also, it pairs nicely with "Doctor". No one would think that a primitive man could be capable of the healing arts. The contradiction of the two words forces one to pause and consider the name.

And that is the hypnotic power of confusion at work to make you consider the usefulness of this blog.

--

Technorati tags: opinion, technology, editorial