Thursday 14 April 2011

A data mining experience

Hands up who went to Spain in August '08? Well, I did with some friends after my mate's wedding. Last night while talking with the said mate, the topic of earthquakes came up somehow via me mentioning Aaron Rolston.

For a little context, I've visited this site a few times recently to see just how many quakes there are around the world every day - turns out a few dozen. Activity around Japan is just phenomenal.

So back in Spain - one morning over breakfast, I remember a friend telling us about a news report about a nearby earthquake that occurred the night before while we were playing poker. He mentioned something at the time, not many of us noticed anything - might have been the San Miguel...

So I thought I'd look to see if I could track it down

This is a five day inclusive search around the world in August 2008 - I repeat, only 5 days worth:

So to find those just in Spain, I restricted the region in the search criteria.

The two in highlighted red in the report below are the ones that we were within possibly 3 kms of it, if I remember our exact location correctly. Two evening quakes just over an hour apart, 7km deep.
                    U.  S.  G E O L O G I C A L  S U R V E Y
                     E A R T H Q U A K E  D A T A  B A S E

 FILE CREATED:  Wed Apr 13 15:52:46 2011
 Geographic Grid Search   Earthquakes=         8
 Latitude:   50.000N  -   30.000N
 Longitude:    10.000E  -    20.000W
 Catalog Used: PDE
 Date Range:   2008/08/04   to    2008/08/08
 Data Selection: Historical & Preliminary Data


 CAT    YEAR  MO DA  ORIG TIME   LAT    LONG  DEP  MAGNITUDE  IEM DTSVNWG DIST
                                                              NFO          km
                                                              TF 
 PDE    2008  08 04 060910.21  42.10   -6.82  17  2.8 LgMDD   ... .......      
 PDE    2008  08 05 152121.24  37.00    3.42   0  4.0 UKMDD   ... .......      
 PDE    2008  08 05 224121.93  38.05   -0.78   7  3.5 MLSTR   3F. .......      
 PDE    2008  08 05 234123.44  38.07   -0.78   7  2.7 LgMDD   3F. .......       
 PDE    2008  08 07 090314.60  42.94   -1.33   2  2.5 MLLDG   ... .......      
 PDE    2008  08 07 103215.30  41.98    2.56   5  2.6 MLSTR   ... .......      
 PDE    2008  08 08 094711.30  42.36    3.45  10  2.5 LgMDD   ... .......      
 PDE    2008  08 08 225611.15  35.58   -4.70  65  2.8 UKMDD   ... .......

I had a look on google maps for the location, I think we stayed in Los Montesinos, Torrevieja - very close!

The internet is full of amazing information - sometimes it's just a matter of tapping into it, sometimes it's a matter of coming up with the idea.

ScottWe

Tuesday 12 April 2011

Observations in pattern matching

Sometimes while driving the suburbs, I'll see a line of cars the stands out for the most benign reason - each boldly coloured - red - blue - yellow - green.

The first time I became aware of how much it stood out for me, I've kept an eye out for it again - a simple alignment of colours. Look around, often you see is really nothing exciting white, white, grey, red, white, white... more variety in the models. So the bright sequence is an outlier that appears very low on my bell curve.

However, I don't really do cars, but when using computers, sometimes I see strange alignments in behaviour. As a developer, I see issues in code behaviour, and sometimes just the way it looks. It's same same but different for testing colour-blindness with the Ishihara color test; and it would be fascinating to hear something with synaesthesia describing the stroop effect, and how the could also be used to find spies!

These days, many people are becoming proficient in using Google. And the engine learns from the use of everyone. People get familiar, particularly with their regular style of searching. I'm reading a fascinating book right now called Click: What Millions of People Are Doing Online and Why it Matters. It's phenomenal the amount of information to be mined & utilised.

That was a tired ramble, this is what I saw that spurred it on
Just like the four words of Isaiah Mustafa "silver fish hand catch", "computers marijuana guns shareholding" were part of four terms I never thought I'd see together in the top four results.

And onya Google for honouring the first human space flight - pairing nicely with another book I've started by Buzz Aldrin.

Over.

ScottWe

Friday 1 April 2011

Of course the Australian Oracle Community is alive

It's just sometimes not everyone hear's about it.

Last Monday I was at the South Australian AUSOUG Conference in Adelaide. This was my first real visit to Adelaide and I was also keen to greet those I've met at previous conferences around Australia, and meet those I haven't had opportunity to see.

The day ended up being quite interesting and fulfilling. SA President Vito Rinaldi mentioned they ended up with around 80 delegates, which he said wasn't bad for this particular event.

As usual, I wasn't overly enthused for the keynote address by Roland Slee, but he had some interesting things to add regarding Oracle in Australia. One interesting point I happened to note down was the notion "economies of scale." It's been a major driving factor of our civilisation for well over one hundred years, but it still doesn't really apply to software development. He took this notion to spruik the power of Oracle Exadata - I was suitably impressed.

I made another great note while in that seminar - an idea for a basic Apex application - don't they come at the strangest times?!

I then had to bust out my first presentation - on Apex Performance. I got some good feedback which was a good offset to the guy I saw at the back of the room nodding off straight after the coffee break. One particular fellow said it was great jot down so many little thoughts on how to generally improve their applications which were basically borne from a self-taught environment.

Next on the agenda was a case study on Agile development by Kevin Wohling. I learned a lot from this one and I thought his analogy of building a house was apt -would you define where to hammer every nail up front? Simply don't over engineer your solutions.
Some good practices were mentioned that I think are regardless of the life-cycle approach - constant communication with users & shield your development team from distractions.

After lunch Guy Harrison's talk on High Performance PL/SQL was his usual high quality. I think info-graphics are great, and his food pyramid for PL/SQL was on the money. From data type usage up the top, down through code structure, data handling and the all important SQL. Again, I noted down another idea - a feature request for the PL/SQL development team regarding bulk binding.

I almost went to Avi Miller's Virtualization and Cloud Computing seminar, but I thought it may have been a little beyond what I was after so I attended Connor's 11g Features for Developers instead. I've seen a variation before, but I'm glad I went as I picked up a few features that I had to note down for looking into further later. I sure wasn't going to soak it all up at the glorious pace that he runs! The most impressive I thought was the ability to manage Editions using TNS services - very cool.
I also noted what he mentioned regarding bulk operations so I didn't repeat the information in my second session.

So my second session was on Bulk Binds, a revision of a presentation I did a few years ago in a WA branch meeting. It was good to leverage what Guy & Connor already glossed over in their presentations and allow the delegates to see a deeper view of what can be done and what should be considered.
The pace was fast, and I'm really thinking the 45 minute timeslots we have here in Australia simply isn't sufficient, particularly when we invite international guests to deliver their 1 hour sessions.

Speaking of international guests, I attended Craig Shallahamer's second session on Evaluating Alternative Performance Solutions. I thought it may have a little outside my league - I was right. I've been happy with the level of mathematics I've seen since high school, and that was a bit much for an afternoon! However he made up for it with his zeal. It was entertaining, and it was a pleasure to catch up with him for drinks afterwards, and ultimately share interesting dinner conversations with him & Tony Scholefield at the Richmond.

Craig has already submitted abstracts for Australia's next event - Insync11 at the Sydney Convention Centre, August 2011. I will be submitting some of my own, so I hope to meet him again in a few months.

Back home in Perth, Sage Computing have some news of our own. We were recently mentioned in an Oracle Customer Snapshot for our work at Department of Treasury and Finance. I can't take any credit for this particular project, but well done Penny, Ray, Eddie & Chris. They all bring aboard specific talents and it sounds like a engaging project - converting a large Oracle Forms application to JDeveloper.

Speaking of, this will be my final segue in this post - Sage have set dates for our next JDeveloper Workshop here in Perth - 13-17th of June 2011 - run by Ace Director Chris Muir.
We will also be scheduling an Apex4.0 course in May or June, which will be an advanced look at Apex4 with special attention to new features - more to come soon.

Phew - I'm glad I'm off to Bali tonight ;-)