Our Blog

JSP compiler configuration in JBoss 4.2.x for ATG

Miro Adamy — Wed, 24 Feb 2010 05:58:42 +0000

By default, the JSP compiler in JBOSS does not generate code compliant with Java 5, even if Java 5 (or Java 6) is the only JVM present.

This leads to exceptions in JSP's that depend on autoboxing feature added by Java 5 - automatic conversion of primitive type (e.g. int) to their object wrapper equivalents (e.g. Integer). As soon as the JSPs depend on this feature, run time error occurs.

How to detect the problem:

If the following page throws exception, your JBoss settings are wrong:

    <%@ taglib uri="http://java.sun.com/jsp/jstl/core" prefix="c"  %>

    DEBUG PAGE

    2+2 ${2+2}

The exception generated is

org.apache.jasper.JasperException: Unable to compile class for JSP: 

An error occurred at line: 6 in the jsp file: /test/dbg1.jsp
The method setValue(Object) in the type SetTag is not applicable for the arguments (int)

3:  DEBUG PAGE
4:  
5:  2+2 ${2+2}
6:  

Stacktrace:
org.apache.jasper.compiler.DefaultErrorHandler.javacError(DefaultErrorHandler.java:92)
org.apache.jasper.compiler.ErrorDispatcher.javacError(ErrorDispatcher.java:330)
org.apache.jasper.compiler.JDTCompiler.generateClass(JDTCompiler.java:415)

org.apache.jasper.compiler.Compiler.compile(Compiler.java:308)
org.apache.jasper.compiler.Compiler.compile(Compiler.java:286)
    ... [[ truncated  ]] ...

If the line 6 is changed to

the JSP file will compile OK.

Error is caused by an attempt to assign int value 10 to variable of type Object

How to fix the error

As described in JBoss Howto the default value for compilerSourceVM and compilerTargetVM is 1.4. We need to set the init parameters to value 1.5.

In JBoss installation, locate the server used for ATG deployment (e.g. atg) and modify the web.xml file in SERVER/deploy/jboss-web.deployer/conf

Add the following init parameters:


    compilerSourceVM
    1.5


    compilerTargetVM
    1.5

Pitfall: Just need to slap the UI on it...

Nael El Shawwa — Fri, 27 Nov 2009 01:26:09 +0000

As much as careful consideration and thought is taken into designing and implementing behind the scenes technologies, the front end is a fairly important layer technology that should also not be underestimated nor overlooked.

Toronto Google Street View Time-Lapse

Nael El Shawwa — Sat, 17 Oct 2009 22:31:00 +0000

Its pretty exciting that Google Street View has finally launched across several Canadian cities.

What is more exciting, is finding myself standing outside the ThinkWrap Solutions office!

View Larger Map

Nicholas Zakas on "Writing maintainable JavaScript"

Nael El Shawwa — Thu, 08 Oct 2009 01:37:00 +0000

A great video on JavaScript best practices by Nicholas Zakas at Yahoo!

Nicholas Zakas: "Maintainable JavaScript" @ Yahoo! Video

I think his best suggestions are:

don't put JS in CSS; its bad for performance as well
don't put CSS in JS; change styles by changing classes, and not actual styles.
don't put HTML in JS; this is rather tricky to avoid, but again if the markup changes, its really time consuming to change it in JS. I guess this one depends on whether you have a lot of markup that gets dynamically updated. One way I really like is how Dojo widgets go about this using HTML templates.
attach event handlers using JS, not within markup ex. "onclick". This is an odd one, but it really does make sense, what if the JS hasn't loaded when the click occurs?
event handlers shouldn't contain business and UI logic; makes a lot of sense since this way I could re-use my event handlers, and I can re-use my business / UI logic
don't compare against "null", compare against the type you actually want.
don't extend objects you don't own ex. Array and extending it with "indexOf" to fix a bug in IE. (Guilty as charged...)
least but not least, throw Errors within functions that can really go bad and you get a useless error like "undefined is undefined". At least now you can tell what went wrong.

Since I have not had enough time to blog about "My thoughts on technology..." I figure I can share videos on others' thoughts on web technology.

Why I use Twitter

Nael El Shawwa — Tue, 18 Aug 2009 00:16:00 +0000

I just read a post by Devin Coldewey about why he doesn't use Twitter. Of course everybody is entitled to their opinion on what they think about a product or service, hence this is my opinion on why Twitter has a place, adds value, and is as complete as a service really needs to be.

Until February/March of this year I think, I didn't use Twitter nor believe in why I would but I then decided to give it a fair try.

Why Tweets have value?

At first, 140 characters may seem too little space to express any single idea, but after using Twitter for a few months I came to realize I really don't need that much more space to express any single thought. Think of it as an elevator pitch. If I can't grab someone's attention in the first 140 characters what makes me think I can in 4 paragraphs?

Twitter in my opinion makes you a better writer and teaches you to express your ideas in the most efficient way possible.

When I scan my feed on my phone or at a computer I'm not really reading every word in the tweet. I quickly scan through and stop when a keyword catches my eye. With a blog I have more text to scan, plus how would I even find the blog post in the first place? And with the shear amount of content on the web, I appreciate the fact that I am harvesting what others found interesting. I might not be interested in everything they tweet, but I followed them for a reason - i.e. I found them more or less interested in the same topics I am.

Twitter is as complete as it needs to be

Sure, people want A+B+C, but nobody said Twitter has to be everything for everybody. Twitter does A really well and leave B and C for others to do just as good. I am a strong believe that if you can't get the basics right, then there is no point to attempt at providing B and C. And when you have the basics nailed, why ruin it by expanding scope? Twitter is not fundamental, but it is simple and gets the job done. I used to scan blogs I follow in the morning before heading to work, now I am exposed to a larger base of articles and tweets where I am able to quickly pick out what I am interested in reading at that point. The problem with following blogs is that most of the time you won't find something interesting on them every day, you will on Twitter.

Why Facebook status updates are nothing like Twitter feeds?

This will probably change after the recent acquisition of FriendFeed by Facebook, but until it does Facebook status updates is very different and is used for different reasons. I would argue that Facebook in fact is pure vanity, I agree Twitter is too to some degree (i.e who has the more followers? who has the coolest background, who is a connector? whose got more RTs? etc.). I wouldn't scan Facebook updates on a daily basis, I would scan Twitter updates though. Twitter is kind of like a dynamic browser and social bookmarking tool. The similarity that I have seen between both is that people are using Facebook to share funny or interesting videos in my case. If I feel like watching a funny video, I would check what has been shared on Facebook.

Another major difference is that on Facebook your interests may not be aligned with your friends as much as they are with the people you follow on Twitter. For example, I got an iPhone a month ago and would like to know what applications people recommend I download. I can easily follow 20 people that talk about the iPhone and find out, how do I find such people on Facebook? and after I do I need to befriend them to be able to see their updates.

Twitter isn't supposed to replace anything, but it definitely adds a lot to certain people - and for me at least it replaced reading online news. Facebook's feed as this stage is nowhere near a competitor to Twitter. Their uses overlap in some areas but in my opinion are disjoint in most areas. Certain things are better shared on Facebook, ex. a picture of friends at your birthday party, other things are better shared on Twitter ex. a link to a specific topic that interests a fraction of your Facebook network, but much more on your Twitter network.

Twitter as a news aggregator

Twitter is a great aggregator, and works great aggregating news which explains why traditional media hopped on it. I didn't learn about Michael Jackson's death on Facebook, nor did I learn about it on CNN, I read it on Twitter. I confirmed it on CNN though and then on Facebook I saw friends' recollections of his greatest moments. On the Iran fiasco, I learned that through Twitter as well, unfortunately that weekend the media dropped the ball and I couldn't confirm any of it, but the fact that so many people were 'reporting' on it in real-time was enough to give it credibility.

Twitter became pure vanity

One thing that disappointed me about Twitter is how fast it got flooded with "experts" and "gurus", there is an expert on every imaginable topic on Twitter - obviously not all legit or credible sources of information. But I am guilty of this as well, I described myself as a "Google Maps Guru".

For me at least, I don't follow many people who just promote themselves i.e. the whole "I just ate an apple....yum" crowd - unless I know them in person or they usually share stuff that interests me. But this is the great thing about its simplicity, if I want to do some sort of social experiment about how often people post when eating an apple, Twitter would be my source for that info.

My daily dosage of tweets come from other Google maps developers, the different Google products teams, friends, legit SEO experts, news aggregators, the media and some miscellaneous people that I found interesting via RTs, FollowFridays or followers.

I agree, our attention is spread so thin these days, but because of that, if it weren't for Twitter's 140 characters could it have worked at all?

Facebook domain type-in hack

Miro Adamy — Fri, 24 Jul 2009 19:00:56 +0000

On Guaranteed SEO Results: Have I got a deal for you

Nael El Shawwa — Thu, 04 Jun 2009 23:52:00 +0000

A search for "guaranteed SEO" on Google returns about 5,300,000 results. On the first page alone 6/10 results mention the word "Google". So this begs the question, is there such a thing as "guaranteed SEO"? Let's find out.

According to Matt Cutts Google ranks pages according to relevance and reputation. Why should you believe him? well because he is the head of the Google Webspam team. When it comes to search; Google Search, Matt Cutts is the man to listen to. So listen to him.

Reason 1: You can't become relevant over night

Being relevant is something you acquire over time. Measuring relevancy is something that gets tweaked over time. So by nature Google's ranking algorithms will and must change over time to maintain relevant rankings and improve their measure of relevancy. So how can you, you or you guarantee relevancy? I can promise you the sun will rise from the east tomorrow (if it didn't then SEO is the last thing on anybody's mind), but can anybody promise you a million dollars tomorrow? Sure someone can; somebody very generous and with a million dollars to give away. For SEO that generous billionaire would be Google, but then if they do, can they remain relevant? You get my point.

Reason 2: You can't become reputable the next day

Building any reputation takes time; even a bad reputation. Your website's reputation is critical for ranking on the SERPs. The more PageRank the higher you rank. You get more PageRank by getting inbound links from others with high PageRank. Quality over quantity. 1 million times zero is still zero.

Reason 3: Because Matt Cutts said so

According to him, the best SEO strategy is adaptive SEO. Just like any optimization algorithm, you make gradual improvements and take a measurement against the objective function. Rinse and repeat, but don't get overly obsessed with it; quit when you hit the point of diminishing returns, then try something else.

This process obviously takes time and research. But thanks to Google they give you all the tools and information you need to rank high on the SERPs.

For my blog I follow three simple rules:

When I write about something, I put myself in the shoes of someone looking for the information I am about to write on. I make a list of these keywords and naturally include them in my post.
Every month I use Wordle.net to generate a tag cloud of my blog. This gives me a good idea about the keyword distribution and whether I have enough keywords on the topics I would like to rank high on.
I make sure each of my posts' URLs contain the main keywords. This step is crucial.

With these three steps, I rank pretty well on some topics:

"photoshop vector cartoons"

"liferay seo best practices"

"liferay virtual hosting"

Now, have I got a DEAL for you... : )

On the GM bankruptcy: Stay small, think big

Nael El Shawwa — Mon, 01 Jun 2009 23:23:00 +0000

Today GM files for bankruptcy protection in an effort to "create a leaner, quicker more customer and completely product-focused company, one that's more cost competitive and has a competitive balance sheet," according to CEO Fritz Henderson. What happened today is only a sequel to many previous episodes where big companies have failed big. Remember Enron? Air Canada? American Airlines? Circuit City? Lehman Brothers? Washington Mutual? WorldCom? Chrysler? Texaco?

Ironically the terms GM CEO Henderson described the future of GM as are "leaner", "quicker", "customer focused", "product focused", and "competitive" which are some of the terms that describe small companies today, in the present not in the future - and after filing for bankruptcy protection.

Too Big To Fail?

Although it is probably applicable for many industries, with regards to technology firms, is small the new big? Why do extra large IT firms lack the creativity and drive to implement the cool applications of today? Why are these cutting edge and innovative applications being created in basements and college dorms nearby by a handful of people? Think Google, Facebook and Twitter for example.

Before the Internet it paid to be big. Big sales team, big headquarters, big advertisements, big, big, big, big and big. Yes I'm talking about economies of scale. Today though, are economies of scale as important as they used to be? Probably not otherwise these big companies such as GM and Enron would not have failed big since they supposedly enjoyed big economies of scale. "Too big to fail" is now a myth of the past. You don't need an army of developers and IT professionals to build successful applications used by millions.

Hyperlocal News - Personalized News For You

Nael El Shawwa — Wed, 27 May 2009 00:01:00 +0000

With online services like Animotto, Flickr, Blogger and Twitter almost anybody can be a video producer, photographer, journalist or news reporter- no degrees or formal training required. With these new mediums the average person can create and distribute their content with virtually no barriers to entry. This major shift gave rise to terms like hyperlocal and citizen journalism.

Hyperlocal is relevant

Hyperlocal news is news relevant to small communities. Think a modern neighborhood flyer or newsprint where anybody can submit content such as videos, pictures, articles or even just their opinions and comments. In the 90s as the internet spread widely, forums and blogs started popping out all over the web to cater to different communities be it virtual or physical ones.

The big news companies took notice of this shift and started providing their own mediums for hyperlocal. These include services like CNN's iReport or The Weather Channel's photo submission. What about the newspapers? how are they coping with these changes and in an uncertain economy?

Once upon a time printed newspapers were extremely popular and they utilized techniques today's websites still use to promote themselves, increase traffic, and keep users on page longer. Sounds familiar? yes these are the same goals every person involved in SEO wants to achieve. Why else do you think that comic strip or crossword puzzle was printed in your favorite newspaper?

Hyperlocal feeds our thirst for news

Newspapers are dying, but people still want news. Our thirst for information has not decreased, if anything it increased beyond belief. The change is, people want to choose what news they want. I never read the stock pages and charts in a printed paper; I rarely read the finance paper and if I do its limited to the front page - so why should it be delivered to me?

The case for hyperlocal news is just getting stronger so perhaps these major newspapers will break off into smaller hyperlocal news outlets. Maybe a visit to NYTimes.com will render a page like NYTimes.com/Jersey? This is if they wake up and stop being in denial.

Hyperlocal is context aware, and of course location aware

Location is a key component for personalization and its possibly the easiest piece of information one can extract from their visitors and deliver personalized content. With hyperlocal news it makes complete sense to me to present it on a map instead of a traditional newspaper-ish view. And with the popularity of mobile devices this strengthens the reasons for a hyperlocal news map. Consider this, you are stuck on a highway and pick up your iPhone to turn on this hyperlocal news map. You get to see where you are right now and what stories are going on around you. The exit is coming up and you are considering escape, but your hyperlocal news map is reporting that a group of chimps have escaped the local zoo and are sitting on the highway barrier 'entertaining' the traffic which is causing all the delay. Now that is something that you don't see on a regular day on the highway, maybe you won't get off that exit after all?

Forget me not, Web2.0 edition

Miro Adamy — Sat, 23 May 2009 16:07:13 +0000