25 December 2011

Atheists will have to do better than this

Christmas Day musings...

A nice little e-flurry has built around several bloggers' trading of a provocative comment from a book by Penn Jillette:

"There is no god and that’s the simple truth. If every trace of any single religion died out and nothing were passed on, it would never be created exactly that way again. There might be some other nonsense in its place, but not that exact nonsense. If all of science were wiped out, it would still be true and someone would find a way to figure it all out again."

Sigh.  Where to begin.

Well, for starters, an exactly parallel construct would be: If every trace of chalk written on a blackboard were to be erased, it could never be written-on that way again.  Which presupposes that whoever wrote on the blackboard wouldn't come and do it over.  Since Jillette's faith holds that there is not and never was a Creator, that makes sense to him.  The chalk, per his faith, was written by men.  And my use of the term "faith" is grounded by Jillette's reliance on such words as "never" and "simple truth."  On inspection, his logic is manifestly circular and self-referential.  (Or is it self-reverential?  [Chuckle] see what I did there).

Meanwhile, science--Jillette's unerring anti-theist lodestone--is hardly canonical across time or place.  For example, dial the calendar back a few years, and a conjecture that gastric ulcers might be caused by microbes would be met by hoots and derision.  How silly!  Bacteria that could survive the acid environment of the upper gut!  There was a day, not long ago, when a researcher proposing such foolish heresy would be laughed out of town, and out of his career.  That's close to what happened to Barry Marshall, who shared the 2005 Nobel Prize in Medicine with J. Robin Warren for that specific apostasy.  There are others: Prusiner and his infectious proteins is another good example.  In the physical sciences, the creative moment documented by the 3K microwave background radiation, so evocative of Genesis, was hugely uncomfortable for scientists to accept.  Years before, Einstein himself had striven to adjust his cosmology specifically to avoid such an event, believing (there's that pesky faith again) that the universe must be unchanging forever (and there's that word, an expression of accepting belief if there ever was one).  As another example, may I mention cold fusion?  The avalanche of mockery that met Pons & Fleischmann's premature publicity crushed what just might have been a spectacular technology in its infancy.  Though the phrase "cold fusion" remains a punchline, much about it remains unexplained, and sober minds are daring to propose a second look.

Here in the nanotech field, the miraculous is observed every day.  As the field advances, each layer of the onion peels away to reveal more onion: more unknowns about nature, and perhaps more unknowables.  Jillette's declaration against the divine is regrettable, as it relegates science to monotonic crank-turning.  On this special day, may we reflect on the scientific value of wonder and awe, and continue proposing foolish heresies.

03 December 2011

No, thank you MAM... did open-source kill Sun?


A few years ago I was given a Sun Ultra 2 3D Creator Sparcstation, vintage 1999, with a then-whopping 1.5GB of RAM, two hard disks and two Sparc processors.

Whoo!  Serious iron.  Weighs a metric ton.  Real hairy-chested UNIX.  Hisses and spits when it runs.

But, it lacked a monitor, and playing with it over RS-232 wasn't too much fun, and I lacked a user account so there wasn't much to do with it.  It sat on a shelf as part of my collection.

Recently I was offered a huge Sun CRT monitor, and it turned out to be compatible.  (And easily 125 pounds.)

So I set it up.

It still left the issue of lacking a user account.  Without system disks, it was impenetrable.  But what I saw was purty...

So I decided to burn a bunch of Solaris 10 update 7 CDs one evening, and start from scratch.  I think the thing had been running Solaris 7 or 8.

Well.  Turns out, with Solaris 10, Sun was deprecating its previous GUI, the Common Desktop Environment (CDE), in favor of Gnome, with which I'm familiar from my Linux usage.  Not a fan of Gnome... so I tried CDE.

And CDE is awful.  It's not what the machine was running before, which seemed airier and more responsive and a whole lot less clunky.

Meanwhile, Gnome is... Gnome.  It's hard to imagine someone spending what this machine originally cost and feeling satisfied with that environment.  It's just nasty.

And despite its great honking 10,000 rpm SCSI disks, two 300MHz 64-bit Sparc processors, and a clock-doubled S-Bus architecture, the thing's a damn slug.

Just abysmal.  I could not be less impressed.

I call it my Mac Appreciation Machine.  Howdy, MAM.

I wonder how the move to Gnome factored into the demise of Sun.  Premium machines, hard-core.  Costly.  Not for home use.  Not for Aunt Min.  Heck, its noises alone would give her the flapping vapors.  No, it's a top-drawer tool for serious professionals.  Yet there it is, glaring at me with the same unpolished face as some crappy netbook running Ubuntu.  Complete with Star Office, seemingly identical to the open-source OpenOffice.

Both Sun and Apple, with OS X's NeXTSTEP-based innards, leveraged the open-source BSD UNIX as their foundations.  In Apple's case, the generic/open-source-y inner UNIX giblets are cloaked with a sublime and solid proprietary user interface with lots of unique and thoughtful goodies built in.  Nothing of the sort with the Sun, at least with Solaris 10u7.  Interface- and usability-wise, I see nothing here I couldn't get from Mandriva or Mint for free, today and maybe even back in 2001 when Sun first started edging towards Gnome, and certainly by 2007-2008 from whence this version of Solaris sprang.

Though it remained (and remains) well regarded in the server space, Sun summarily disappeared from desktop usage, and I wonder if Gnome was a symptom or a cause.

My thought: as in every business endeavor, differentiation is everything.  Whatever other problems Sun was battling in the market, it also lined up a chunk of its differentiation carefully in the cross-hairs and blew it away by adopting an open-source persona for its machines.

It seems my thoughts both parallel and oppose those of Scott McNealy from exactly a year ago.  On the one hand, the interviewer refers to the "open core" of Sun's products, which could have excluded the user interface.  And McNealy notes, "We probably got a little too aggressive near the end and probably open sourced too much and tried too hard to appease the community and tried too hard to share... You gotta strike a proper balance between sharing and building the community and then monetizing the work that you do... I think we got the donate part right, I don't think we got the monetize part right."

But he doesn't mention differentiation.  And if McNealy & Co. were prescient in stating that The Network Is The Computer, maybe they missed appreciating that The Interface Is The User Experience.  And on MAM, with Gnome, that's nothing special.