L.A.B. Unleashed

On NPR’s Marketplace segment last night, they were talking about Microsoft [finally] delivering Vista to the consumer today.  Microsoft tried to hype things up a bit by making copies of Vista available at 12:00am today, Tuesday 1/30/2007, similar to stunts pulled for Harry Potter books or Star Wars movies when they hit the market.  The more interesting part of the segment was an interview with Scott Rosenburg about his new book, Dreaming in Code: Two Dozen Programmers, Three Years, 4,732 Bugs, and One Quest for Transcendent Software.  Rosenburg’s quotes about programming large pieces of software, such as Notes or Domino or Vista, were right on.  His thoughts on measuring development’s progress ….

…sometimes people in the field try to measure it by lines of code, it’s sort of like how many words have you written on an article. And the problem is that sometimes the best piece of code is one that uses very few lines to accomplish what it’s trying to do. So it’s very hard to measure progress…

I’m intrigued by the book.  The title is key catching, and I always say I get my best ideas for solving coding problems in my sleep.  The reviews on amazon.com allude to connections to the development of Lotus 1-2-3, and that the book does dig deep into some OO programming discussions.  Going to check Borders for it today, I’ll let you know how it is ….

Lotusphere 2007 wrapped up on Thursday afternoon, and it was a really great week.  The lab was pretty busy and my talk on Domino, Websphere, & Portal Administration integration went really well.  Highlights were definitely talking to customers in the lab and I appreciated all the feedback they gave.  I spent a good portion of my day today putting in enhancement requests … we had some excellent ones.  Not to mention a few pesky bugs I’d like to see fixed.  Also, in the lab, we were trying to get customers to take our Administration Survey, as the responses from it is really going to influence what we do for Domino next.  Our first completed survey was returned by one of my favorite customers I met all week.  He was native to Mexico, but worked in Georgia.  After answering his questions about DDM and how to change event severities, I mentioned that I have been learning Spanish.  We really hit it off, and the next day he returned our very first completely filled out *long* version of the Admin Survey!  We were all very appreciative!  For more information on Domino 8 Administration and if you would like to get a copy of the survery, please post a comment here with your email address.

I knew I wasn’t going to have time to blog all week.  It was a whirlwind, as I expected, and any free time I had, I tried to catch up on some sleep and rest my brain.  Kevin came down on Thursday afternoon and we spent the weekend at the Dolphin.  Jelly Rolls was a blast, and we also enjoyed a couple of nice meals at Shula’s [awesome filet] and Kimono’s [yummmmy sushi].  After raining all week, the weather warmed up on Friday and Saturday making for a great round of golf and time to relax and enjoy the outdoors.

I have to admit, today was a little rough returning to the 25 degree weather, with windchills in the single digits.  I was happy to get home to the meat head though, and he was happy to see that I brought him a couple gifts.  I ended up getting an extra one of the highly coveted yellow "cult" shirts.  Good thing it was Kobe’s size!

 

I am really looking forward to getting on the road and heading to Orlando for my second Lotusphere experience!  I am leaving just in time too, because this is the current state of my kitchen …..

I’m not sure I’ll get much blogging in while I am down there, it is always a whirlwind, but I will try.  I better head out now, because my talk is on Tuesday morning at 8:30am.  Even though it’s a straight shot down I-95, it could take some time on the scooter ….

Makers of magnetic disk drives in Boston are realizing that as the demand for more storage increases, they are headed for a road block if they continue building drives with their current technology.  An article in Science discusses the current model of boosting disk drive storage space is to shrink the regions of the spinner platter, in order to fit more bits on the drive.  It is predicted that this model will continue until around 2010, when disk drives are expected to pack 500 gigabytes per square inch.  According to the article, this is equivalent to a laptop storage device of 320 to 640 Gbs.  The major contender to take over the current technology is patterned media ….

In patterned media each bit is made from a single magnetic island that’s not broken into smaller grains. Because these islands can be larger than single grains, they are less apt to be influenced by ambient temperature fluctuations and thus are more stable.  That in turn should allow diskmakers to pack those islands closer together than the bits made from bunches of grains today.

The standard patterning technology of the semiconductor industry, known as photolithography, can now pattern features down to 65 nanometers.  But to beat a data density of 500 Gb/in2, diskmakers will have to make their magnetic islands a mere 22 nanometers across with just 36 nanometers between the bits.

This technology is not only expensive, there are many different ways to implement photlithography, and industry experts have not agreed which way to go yet. 

I did a quick search to see what the largest laptop harddrive for sale is currently, and came up with a 200 Gb one by Toshiba.  My IT guy says Seagate has put out a 750 Gb one that is 2.5 inches, fitting into a laptop.  From this article, it looks like Seagate boosted its drive capacity in a different way than patterned media.

The big leap stems from a new so-called "perpendicular recording" technology that allows drive makers like Seagate and rival Hitachi Global Storage Technologies to boost the density of a disk by aligning bits of data vertically rather than horizontally. At the same time, fewer moving components are needed in the drives.

Perpendicular recording was also discussed in the Science article as a more efficient way to orientate the bits on the platter. 

It seems like data storage is going to be a hot topic in the near future, and that as our demand for video, mp3, and picture storage increases, the industry is going to need to get creative to solve these problems.  It has taken 50 years to reach this plateau, as IBM produced the first storage devices in 1956.  With a little help from our friend, nanotechnology, it looks like the future will be bright in this area!  What an exciting time to be a part of this industry!

There has been no argument for some time, that the most efficient way to stack spherical objects, such as cannonballs, is in the shape of a pyramid.  Seems logical, almost obvious, right?  There are many examples of this everywhere you look – produce stacked at the supermarket, trail markers at the top of mountains, and of course I am sure you are always surrounded by cannonballs stacked on the deck of a boat … who isn’t really?! 

Well this is a mathematical problem that has remained unproven since 1611, when Johannes Kepler proposed the conjecture, but was unable to prove it.  Just under 400 years later, we finally have a proof, and that is only because we now have the computer power to actually do the computation.  Thomas C. Hales, a University of Pittsburgh Professor, and a former graduate student Samuel P. Ferguson, published their proof in 2006, albeit they actually solved the proof 8 years earlier, but it took all of that time for a panel of 12 mathematicians to confirm there were no mistakes.  So, 300 pages, 40,000 lines of computer code, and 3 billion bytes of data later, Hales and Ferguson reached a monumental accomplishment. 

He and Ferguson studied 5,000 configurations of stacked spheres, which they reduced to 100 candidate configurations.

Then after years of effort, the "eureka moment" occurred in November 1994, when Hales figured out the ideal geometric forms that best described the relationship between spheres and the space they occupied.

Reducing the problem into creative geometry allowed a computer to do the calculations. But the computer had to run nonstop for three months to do many billions of calculations to complete the proof, he said.

Can you imagine letting a computer chug on one problem for 3 months?!  You’ve got to be ecstatic that the machine didn’t crash in that time, let alone actually give you the answers you were looking for!  Computers have been so instrumental in the advancement of mathematics and so many other fields as well.  Another theorem that comes to mind is the Four Colour Theorem, which I have discussed before here.  It makes me wonder, how much further technology will need to be developed, until we see some of the "unsolvable" mathematical problems solved.  P vs. NP for instance … what kind of technology is it going to take to solve that one …. and how many computers will crash trying?

Link to full article: http://www.shns.com/shns/g_index2.cfm?action=detail&pk=MATHWHIZ-01-16-07

This is hilarious …. Get your taste now, because it conflicts with Grey’s Anatomy & The O.C.  I know I know … I AM ridiculous … but there is only so much the DVR can handle, and with The O.C. going OFF the air, I’ve got to be loyal until the very pathetic end.

I have been awaiting my specially ordered copy of the latest Science journal, the one that has the breakthrough of Poincare’ on the cover.  After ordering the issue online, it took two weeks to get to my door, but luckily it was worth the wait.  The cover alone is so exciting to me, as it is a beautifully simple representation of the mathematical science of topology.

Inside, there are tons of other interesting articles, and one that caught my eye was entitled, A Scientist’s Nightmare: Software Problem Leads to Five Retractions.  Basically, the article is about a scientist who published findings produced by faulty software.  I’ve learned from friends who are scientists that to get results, sometimes they do a lot of their own programming in C++ or java.  So, not only are they experts in their own fields, they need to be programmers too, to analyze the data that they research.  Apparently, in this case, the software was passed down from lab to lab, and the bug (where 2 columns of data were being inverted) was never fixed.  From the results of the software, 5 separate articles were published, and many others based their research off these articles.  Oh the horror!  I totally feel this guy’s pain.  It reminds me of the times I have broken the daily build, or have found a bug in the new feature just released.  That moment that your realize there is a problem, sheer panic strikes.  BUT … the moment you realize WHAT is CAUSING the problem, there is a sense of relief and confidence that it can be fixed. 

Interestingly enough, the data is making a lot more sense since the bug has been fixed in the piece of software.

Chang meanwhile has been reanalyzing his original data and expects to submit papers on the corrected structures soon.  The new structures "make a ton of sense" biologically, he says.  "A lot of things we couldn’t figure out before are very clear."

And that’s the other thing about making a big (or even small) bug fix, everything just falls into place and makes so much more sense after those nasty little bugs are gone!

And don’t even think about cutting him short by a day or two!  My Dad’s retirement ceremony was held Friday at Hanscom Air Force Base.  There was an overwhelming amount of people there, and it was amazing to see all those people admire and respect my father just as much as we all do!

 

Here’s to a happy retirement filled with long drives and short puts!

Last April, I posted an entry about how Kevin was hounding me as the King-Of-Technology in the house.  Well, nine month later, I think this proves there is a new Queen in town … what’s up now?!  Where’s my paper-mashay crown?