Wednesday, Dec. 14, 2005 (A.G): A Rare Book Review -
"City of Liars" by Roger Willcocks

No, this is not a story about D.C. – and just in case some of you who are not that conversant with good ol’ Anglo-Saxon names are wondering, that is also not a nom de plume.  So you will not find any insights into Scooter or Karl in this book, though, alas, you will encounter a ‘Tonic-Head’ (sic), which might aggravate some of you given its overtly racial type-casting (though you could, if its helps, chuckle over it given that I probably looked like that 20 years ago, when I too played cricket at every opportunity that I had).  OK, so that was a clue.  The City here, in every sense of the word, is the legendary, gilt-edged square mile at the very heart of London (Great Britain as opposed to the, at a minimum, 10 you can find in the U.S.).

Though a thriller, set against a backdrop of high tech and finance (as the cover states), this is not a Grisham nor a Dan B. – though I actually found that reassuring because Dan’s first ‘non-tech’ thriller, in 2000, despite being rather good (and in a very, very different league to this book) also had a few, distracting, inaccuracies.  Thus, with luck, Roger like Dan may outdo himself with his next book, in which case we can all say: “Ah!  I was familiar with Willcocks before he became a mega star”.

So why am I, so atypically, bringing this book to your attention – other than the fact that I owe Roger a few favors from a long time ago (one of them, and probably the most germane, being for the fact that he told about Annie Proulx’s mesmerizing (not to mention Pulitzer winning) book “The Shipping News”).

The real crux of the matter, however, is that Roger used to really be one of us … and to an extent probably still is though he is spending more time as a writer.

Roger, though you will not find this in his bio that appears in the book, used to be the President(/G.M.) of Intelligent Environments (iE) that among other things used to have one the neatest (and as such one of my favorite), ‘on-the-fly’ host-to-HTML products – iE ScreenSurfer.  So that is the hook.  Roger, indubitably, was an high-tech, dot-com initiate.  Thus it would be ungracious of us not to give this book a plug.  Hence this write-up.

Given the tacit pledge that I have with you all for honesty, veracity and integrity, I do have to (blushingly) confess that this book does have some issues pertaining to continuity, credibility and overall plot.  But that in a way makes it an intriguing and worthwhile read.  Make that a challenge.  Set out to find the anomalies.  Though Roger, as yet, has not instituted such a program, lets assume that Roger, like the fabled Knuth will reward you with a small cheque, or to be more ‘hip’ with a small PayPal credit, when you notify him of sustainable issues that he really should fix in the next issue.  [Somewhere around 1972 I did get a Knuth check, though I was not even aware of his ‘program’.  I wrote him a letter, as was the then norm, telling him that I just could not get one of his algorithms to work.  About six months later, I just got a cheque, but not an explanation as to how to fix the algorithm!].

You can get the book from Amazon UK (but not Amazon.com).  So click here to place the order.

So here is what I propose.  This book will be an ‘OK’ read if you are on a trans-continental or trans-Atlantic flight, especially if you have made sure that you have availed yourself to an adequate supply of booze.  The protagonist in this book are rather partial to their booze (and some casual sex), and I, without compunction, will recommend the former activity as a good way to ensure that you do enjoy this book.  So get this book, read it on a plane (not forgetting the booze) as and maybe drop Roger a note – so that you may try get a modicum of reimbursement for the refreshments consumed.

The bottom line, however, is that I tip my hat to Roger.  It takes a lot of guts to write fiction.  I unashamedly admire all that have the courage and conviction to do so (including my father, who turned his hand to fiction in his late 60s, having prior to that penned upwards of 40 non-fiction books).  So, Roger: “Cheers” … happy New Year … and with Dan as your inspiration, I hope that the next one is going to break the code.

Thursday, Dec. 08, 2005 (A.G): EE vs. CCL vis-a-vis IBM 37xx Replacement

An 'advertorial'-oriented, print media, mainframe publication dated ‘October/November 2005’ contained an article entitled “Alternatives for Replacing IBM’s 37x5 for z/OS Users: A Comparison – CCL or Enterprise Extender?" -- as some of you, alak, may have seen (and possibly even read).  Since we have become the de facto home site for the IBM CCL, with #1 ranking from Google (which is appropriate given my legacy as the erstwhile Mr. SNA), it would be remiss of me, if I did not point out some of the inaccuracies (if not all of the amusing off spin) in this article.  To that end, my BIG Iron club members already received an exclusive, somewhat technical e-mail this morning -- per our charter.

The main problem with this article is that the CCL functionality and performance quoted in this article pertains to CCL V1R1 – as was available in March 2005.  As I know from experience, this is the perennial problem of writing about 3rd party products for a print publication, especially when it has extremely long lead times.  Hence my penchant for this, real-time media.

CCL V1R2 was announced on October 24, 2005 and was generally available as of November 2005.  So this article was out-of-date even before it saw the light of day!

If you haven’t downloaded it as yet, PLEASE refer to my (IBM approved) CCL/z V1R2 White Paper [600KB PDF; 20 pages].

It is also available at www.ibm.com.

Refer to Page 6 for the “What is New in CCL/z V1 R2”.  Also check out the exclusive, 2-page table on “CCL/z V1R2: What is Possible, What is Not” on pages 10-11.

Based on this you will realize that some key ‘claims’ made in the article are no longer valid.

If you want MY take on this, join the BIG Iron club ... and then drop me a line.  OK?

With luck, I plan to also have an IBM interview on this germane topical issue before Christmas.  At that juncture we will also look at how mainframe DLSw, to be available with CCL in 2006 (as discussed in my White Paper and consistent with IBM's SoDs), adds another dimension to the EE vs. CCL/z V1R2 selection criteria matrix.

Friday, Sept. 16, 2005 (A.G): z/Linux Has To Overcome The Linux/Intel Mindset

With IBM continuing to relentlessly push Linux, with (anecdotal) claims of customers flocking to it and sponsored research papers that purport that Linux is significantly more cost effective than Sun Solaris, I did a bit of my own (haphazard) Linux related research when I went to SHARE (in Boston), last month, to meet a few people and see what was afoot, in general, in the IBM world.  I managed, as usual, to gleam some interesting (but little discussed) insights, particularly related to the much vaunted platform independence of Linux -- as exemplified by IBM's rather attractive Linux solutions on mainframes and the POWER5 machines [i.e. p5s and i5s].  While folks in the IBM world fully appreciate that Linux is available on a plethora of disparate platforms, that spans all the servers that IBM peddles and even Sun servers, it appears that many corporations, particularly in the financial sector, have decided, at least for the time being, to kind of ignore this!  Yes, you 'heard' me right.

Though true platform independence is one of the many beauties of Linux (and something that indisputably sets it apart from Windows 2003), many IT organizations appear to have decided that this is just a distraction.  Linux, implicitly, is being equated with Intel servers.  Yikes.

There is nothing wrong with Linux on Intel, and we all know that Linux/Intel is a proven, highly resilient, cost-effective server configurations.  But to dictate that Linux has to always go hand-in-hand with Intel, is like saying that you will only eat beef in the form of McDonald's Big Macs.  You are, at a stroke, precluding too many other potentially attractive options.

But this is reality.  So while IBM talks about the possibility of running 40 'zillion' Linux images on a mainframe, big banks, though committed to mainframes, are making tactical decisions based on the edict that Linux is fine as long as you are talking about Intel servers.  I was in luck.  I was having dinner, that night, with a key (maybe THE key) z/Linux guru from IBM.  We talked about this platform issue.  As was to be expected, IBM kind of knew about it, but wasn't sure how to really tackle it head on.

So that was the first interesting thing I learned about Linux in the IBM world. But it wasn't all. Also learned something nearly as interesting about what folks are considering to be the hidden costs of migrating to Linux from Unix.  I will cover that in another BLOG entry soon.  Promise.