Are your daily stand-up meetings dull and boring? It doesn't have to be. Each person in our team do our level best to bring some fun into the meeting.

Kevin started the "When is next Tuesday?" debate 174 days ago. To this day, any mention of any dates next week would bring the whole joke up, culminating in Rob's "but this week was next week last week."

Last week Rob brought up the topic of English spelling reform. I recall seeing something similar many years ago. Today I went into my stack of printed out stuff and found it. A little Googling lead me to this jem:

> Here's a bit of fluff to inject some humor into these trying days in The
> Place:

> Having chosen English as the preferred language in the EEC, the
> European Parliament has commissioned a feasability study in ways of
> improving efficiency in communications betwen Government
> departments.

> European officials have often pointed out that English spelling is
> unnecessary difficult; for example: cough, plough, rough, through
> and thorough. What is clearly needed is a phased programme of
> changes to iron out these anomalies. The programme would, of
> course, be administered by a committee staff at top level by
> participating nations.

> In the first year, for example, the committee would suggest using
> 's' instead of the soft 'c'. Sertainly, sivil servants in all
> sities would resieve this news with joy. Then the hard 'c' could be
> replaced by 'k' sinse both letters are pronounsed alike. Not only
> would this klear up konfusion in the minds of klerikal workers, but
> typewriters kould be made with one less letter.

> There would be growing enthousiasm when in the sekond year, it was
> anounsed that the troublesome 'ph' would henseforth be written 'f'.
> This would make words like 'fotograf' twenty per sent shorter in
> print.

> In the third year, publik akseptanse of the new spelling kan be
> expekted to reatsh the stage where more komplikated tshanges are
> possible. Governments would enkourage the removal of double letters
> which have always been a deterent to akurate speling.

> We would al agre that the horible mes of silent 'e's in the languag
> is disgrasful. Therefor we kould drop thes and kontinu to read and
> writ as though nothing had hapend. By this tim it would be four
> years sins the skem began and peopl would be reseptive to steps
> sutsh as replasing 'th' by 'z'. Perhaps zen ze funktion of 'w'
> kould be taken on by 'v', vitsh is, after al, half a 'w'. Shortly
> after zis, ze unesesary 'o kould be dropd from words kontaining
> 'ou'. Similar arguments vud of kors be aplid to ozer kombinations
> of leters.

> Kontinuing  zis proses yer after yer, ve vud eventuli hav a reli
> sensibl riten styl. After tventi yers zer vud be no mor trublsm
> difikultis and evrivun vud fin it ezi tu understand ech ozer. Ze
> drems of Mr. Orvel vud finali hav kum tru.

> From REA News, Journal of the Royal Aircraft Establishment. 

Responses[1]

Posted by Weiqi Gao on January 30, 2006 10:05:55 PM CST #

Tomorrow, January 29, 2006, is Chinese New Year, the Year of the Dog. However, North America being 13 hours behind China, it is already January 29, 2006 there. To all my Chinese and Asian friends, a Happy and Prosperous New Year.

To continue the tradition that I started last year, 353 days ago (remember the lunar year is 354 days), here's some more Chinese trivia:

• We'll have a leap month this year. The eighth month will be called the leap seventh month. It runs from August 24 till September 21. It misses the autumnal equinox, which is on September 23, by two days.
• In traditional Chinese age counting (虚岁), a new born baby is one year old. And everybody's age gets incremented by one on Chinese New Year. So a baby born a day before Chinese New Year could be two years old a couple of days after birth.
• The animals are not officially part of the calendar. It is a way to ease the memorization of the official year name, which is an interesting numbering system in itself. Let me explain,

The Chinese year has a two character name. The first one cycles in a set of ten stems: 甲, 乙, 丙, 丁, 戊, 己, 庚, 辛, 壬, 癸. The second one cycles in a set of twelve branches: 子, 丑, 寅, 卯, 辰, 巳, 午, 未, 申, 酉, 戌, 亥. And the trick is that on the Chinese New Year, both characters are incremented. Thus the new year is 丙戌, while the year just ended is 乙酉.

Since lcm(10, 12) = 60, the two character year names cycle through every 60 years. As you can imagine, seeing the beginning of a new cycle (the year 甲子) is an exciting event, akin to seeing the beginning of a new century in western calendar. The cycle being 60 instead of 100 means that almost everyone get a chance to experience it. Very pragmatic. (But don't try to start a JSR for java.util.ChineseCalendar.)

Historical events are recorded in the two character year names, usually coupled with the dynasty name and the emperor's era name for disambiguation. The most important events are usually recognized by the two character name only. 甲午, for example, almost always refer to the First Sino-Japanese War of 1894. Similarly, 庚子 almost always refer to the 1900 Boxer Rebellion. 甲申 marks the fall of the Ming Dynasty in 1644, and 辛亥 marks the 1911 revolution that forced the Last Emperor of Qing Dynasty (1644-1912) to abdicate the next year, at the age of 6.

Add a comment

Posted by Weiqi Gao on January 28, 2006 3:50:31 PM CST #

I spotted this message on the GNU Classpath mailing list yesterday:

Michael Kay: I'm working on a port of the Saxon XSLT and XQuery processor to the .NET platform using IKVMC. (You may already be familiar with this as Saxon.NET, but I'm now looking at folding it into the core product).

I've followed the Free Java (GNU gcj, Apache Harmony, GNU Classpath), .NET (Mono), and XML (XQuery, XSLT) spaces separately for a long time now. Today's post links the three together in a uniquely surprising yet logical fashion.

For people who don't know, Michael Kay is editor of W3C XSLT 2.0 and XPath 2.0 standards, and author of the popular open source XSLT/XQuery implementation Saxon. IKVMC is the Java bytecode to .NET CIL compiler from the IKVM project by Jeroen Frijters. It turns jar files into .NET assemblies for use under both Microsoft .Net Framework and Mono. It uses GNU Classpath as its underlying class library.

I think this move sends several signals to the wider Java community, especially those who dismissed the GNU Classpath effort when it started in the last millennium as a waste of time and have not revisited it since:

1. GNU Classpath is more complete than its 0.20 designation suggests. Most non-Swing Java 1.4 applications libraries should be runnable with it.
2. GNU Classpath based compilers and runtimes have already fulfilled their potential of taking Java to new and unexpected places.
3. It pays to make sure that your library works with both the Sun JDK and free Java because it will take you places you never thought you could go.

Take a look at the free tools. You may have already won a free ticket to those places (natively compiled executables, .NET environment, bundled with a Linux distributions and end up on millions of computers)! if your code compiles and runs with a free compiler and runtime.

Responses[1]

Posted by Weiqi Gao on January 28, 2006 12:42:01 AM CST #

I had a lunch time chat with Ken Totten, who pointed out something interesting on OCI's own website (I'm so used to reading everybody else's website, I ignored my own employer's):

Object Computing, Inc.: Separation by time, distance and culture increases risk to already technically challenging activities. Management should be trying to reduce risk, not magnify it. Failure, even at half the previous cost, is not a management option.

The Offshore Technology Audit check list is also worth a read.

Add a comment

Posted by Weiqi Gao on January 27, 2006 6:39:10 PM CST #

(Via Andy's Blog)

The Waterfall Alliance: After years of being disparaged by some in the software development community, the waterfall process is back with a vengeance. You've always known a good waterfall-based process is the right way to develop software projects. Come to the Waterfall 2006 conference and see how a sequential development process can benefit your next project. Learn how slow, deliberate handoffs (with signatures!) between groups can slow the rate of change on any project so that development teams have more time to spend on anticipating user needs through big, upfront design.

Add a comment

Posted by Weiqi Gao on January 26, 2006 4:50:59 PM CST #

Just noticed this:

Yes. The delete button. Finally.

Responses[2]

Posted by Weiqi Gao on January 19, 2006 8:37:12 PM CST #

Following Mike Shoemaker and Matthew Porter’s lead, I have implemented Glen Smith’s Maths Captcha patch for Pebble 1.9.

Pebble’s spam blocking capabilities is already pretty strong (I cleaned up more than 300 spam comments in the pending bin yesterday and today.) Yet there still are a few that made it past the spam filters. Since I have the “email earlier commenters when new comments are added” feature turned on, quite a few spam emails were sent out. Eric told me he receives them occasionally. I feel really bad about it and apologize to anyone who might have received such spam emails.

Here’s a big Thank You to Glen.

Responses[2]

Posted by Weiqi Gao on January 14, 2006 7:16:45 PM CST #

Go read it and laugh.

Add a comment

Posted by Weiqi Gao on January 14, 2006 8:23:42 AM CST #

The following exchange happened at work today:

Forget about the cool stuff that I mentioned for a moment, and let’s talk about the abuse that Cygwin users take in situations like this. I’m used to it and usually go with the flow when it comes to opening a CMD prompt. However in all these years of opening up a CMD window and redoing stuff, not once had the problem been traced back to Cygwin.

I don’t mean to pick on Dan. (Others do the same.) I just want to point out the not so obvious fact that Cygwin is rarely the cause of any problems on my machine.

Responses[3]

Posted by Weiqi Gao on January 13, 2006 7:15:22 PM CST #

Today’s talk at the St. Louis JUG is on HttpUnit, by Lauri Peterson from Technology Partners. Lauri had a lot of real world experience with HttpUnit and talked not only about the stuff that you can learn from the tutorials, but also the pitfalls and irregularities.

Today’s quote of the day, however, is not about HttpUnit. It’s from Lauri’s opening ice-breaker story, which has her getting a recruiter phone call at diner time: “I’m Richard So-and-so. I work for Google. I was searching the web and your name came up. ...” Then she realized:

“I’ve been Googled by Google!”

Add a comment

Posted by Weiqi Gao on January 12, 2006 8:57:57 PM CST #

Mark Volkmann asked a simple question on our internal Java mailing list: Are Java Thread priorities being ignored on the Linux platform (Fedora Core 4 on i386 with Sun's latest JDK 5.0 update)?

I can't seem to find the answer in a definitive way.

I would appreciate it if someone who has gone through researching the answer to this question give me a hint.

Thanks,
--
Weiqi

Responses[3]

Posted by Weiqi Gao on January 12, 2006 9:32:11 AM CST #

The following two pieces of news will probably prove more important in the future than they seem right now.

Miguel de Icaza: I just woke up to the announcement over at Chris Blizzard's blog (from Planet Gnome) that Mono today was included in Rawhide, the staging area for the next release of Fedora Core.

Tom Tromey: Classpath hit 98% of 1.4 today, and I think we'll see another big bump tomorrow since the XMLEncoder patch went in today.

Add a comment

Posted by Weiqi Gao on January 10, 2006 8:16:53 AM CST #

(Via Ted Neward)

Chris Sells: This is how you tell the marketing guys from the engineers... : )

Responses[11]

Posted by Weiqi Gao on January 4, 2006 6:50:14 PM CST #

Mark Volkmann: Using SWIG reduces the amount of manual coding required to invoke C/C++ functions from other programming languages. In the case Java, JNI can be used directly, but this requires a large amount of tedious coding. Another benefit of using SWIG is that it is less error prone than manual coding. Correct use of JNI is complicated, especially for passing non-primitive types.

I think SWIG is something that is underreported and underutilized in the Java community. And it is a very useful thing. It promotes reuse and cuts down on the number of wheels that you have to reinvent.

Aside from a couple of minor problems caused by smart dashes and location differences between i386 and x86_64, the sample run fine, both the Java and the Ruby version, on my newly built Fedora Core 4 AMD64 box:

[weiqi@gao] $ ./running-swig-for-java
[weiqi@gao] $ ./running-the-java-application
Mark Volkmann is 44 years old.
Amanda Volkmann is 20 years old.
Mark Volkmann is older than Amanda Volkmann? true

And I haven't even installed the Sun JDK yet. ( Fedora Core 4 users are advised not to use the Java RPM provided by Sun.) I'm running with the Free Java (a mix of gcj, GNU Classpath, and Eclipse) that comes on the Fedora Core 4 CDs.

Responses[1]

Posted by Weiqi Gao on January 2, 2006 9:17:11 AM CST #