Today I discovered that a backup routine I thought existed was a figment of my imagination. Nothing was lost, nobody died, but it put a shiver up my spine around what might have been--or not been. The necessary script was quickly cobbled together and tested, so all is well. The gap was small and affected only my personal websites: those I run for others have been backing up nightly forever.

The episode gave me pause for thought. How many of us have lost important work through failing to save a file, or to back up important work? Murphy knows. Yet why is it so difficult: why do we fail to do it so often?

The truth is, it's a chore, and the busier we are the more of a chore it is. Programs that do automatic file saves are a godsend, provided they do so without one having to set a preference first. Backing up whole filesystems, webservers, mailboxes and the like is another matter. Where are the operating systems that do this automatically, without one even having to think about it? For sure, there is (expensive) software that will do the trick, but it still takes time and effort to get it working.

One of the allures of Mac OS X is its Unix underpinning which should, in theory, make backing up a relatively easy and inexpensive business. After all, tape archiving (tar) is at least as old as Unix itself. But as it turns out, its not quite so straightforward: the Mac HFS filing system has quirks (resource forks and the like) which can behave badly with the traditional Unix routines such as cp and the like. Slowly, Apple is modifying these routines to behave more coherently with it's filing system conventions, but its still a minefield. And to cap it all, the lovely LaCie mini firewire external hard drive that sits under the Mac mini on my network as a backup device is quirky about setting owner and permissions on its folders.

All said, backups are happening here nightly. So far, Murphy has been benign and I haven't had to retrieve anything in anger. Testing a the restorability of a backup is something I suspect is rarely done until its too late. I can't say I'm squeaky clean on that one either: it all takes time. The trouble with this risk management business is that it's a never-ending story--even before you think about the precautionary principle.


It is only proper that a weblog with the name Jambalaya! should pause and reflect on the devastation that Hurricane Katrina has wreaked on the home of the said dish, the loss of life, and the virtual destruction of its major city, New Orleans.

I have visited New Orleans twice and have very fond memories. Although it has its problems, it was a fun place to be, and no human community deserves the fate that has befallen it. Our thoughts at Bayswater Farm are with you all.

Another visit was looming this December, for the American Society of Hematology annual convention. My first visit to both New Orleans and ASH was for such a convention twenty years ago. It was a large event then, but would have been massive this year. It seems from this side of the Pond that there is no way this year’s convention can now take place.

New Orleans thrived on tourism, and conventions such as ours will have been major players in its economy. Should this year’s ASH convention not now go ahead, I suggest that the least we could do is donate our registration fees to a relief fund for the people of this stricken city.

(Addendum: It now seems that ASH plans to relocate its annual meeting to another city. I may or may not go, but have made a donation to the Red Cross.)

Cetriolo—cucumber for Italians

Skip Lombardi has issued a plea on his Italian Food Blog for Italian recipes containing cucumber, which he has hardly ever seen in Italian cuisine. It had never occurred to me before that cucumber has made such a minimal impression on the Italian palate. Perhaps it has been eclipsed by its cousin the courgette, or zucchini. One doesn’t see many zucchini sandwiches: unlike cucumber, which is purely a salad vegetable, zucchini generally has to be cooked.

As Skip points out, cucumber is a key ingredient in the famous Tuscan salad, panzanella. And salad it is with the few other cucumber recipes I have come across. Marcella Hazan, in The Essentials of Italian Cooking has recipes for orange and cucumber salad, and roasted aubergine with peppers and cucumber. Elizabeth David, in Italian Food, has an intriguing Insalata di finocchi e cetrioli. I guess these could all be classed as types of insalata mista. Cucumber-only recipes however are rare: the only two I have come across are from Janet Ross’s early 20th century Leaves from our Tuscan Kitchen, more recently re-edited by her great-great-nephew, Michael Waterfield:

Cetriola alla comasco. Cut some strips of peel from a cucumber and slice the cucumber very fine on a mandoline. Arrange on a dish and sprinkle over half an onion (grated or very finely chopped), 1 tbsp tarragon vinegar, and 2 tbsp good olive oil. Allow to ‘pickle’ for 15min before serving.

Cetriolo condito al miele. Cut some strips from a cucumber, cut the cucumber into inch pieces and then into rather thin wedges. Pour over the following dressing: 1 full tsp honey, salt, pepper, pinch chopped marjoram, 2 tbsp wine vinegar, 4tbsp olive oil.

Connecting for Health: too little too late?

A rather depressing article in E-Health Insider reports on what sounds a bit like a panic measure:

NHS Connecting for Health is urgently looking for experienced NHS clinicians, interested in ensuring that NHS IT systems are fit for purpose, to be flown out and work on assignment with clinical software developers in India or the United States.

It has been my impression for some years that the politicians driving this (not very original) vision have been blindly seduced by BigCo IT into believing that simply rolling legacy systems from across the Pond and imposing them on an unwitting NHS would be a piece of cake. Not so. As any observant student of large IT projects will tell you, a system stands or falls on its usability far more than its underlying sophistication, and you cannot assume that usability in one context translates simply into usability in another. Put more simply, the North American healthcare system and the NHS differ in so many social, cultural, and professional respects that a system designed for one will almost certainly fail in the other without significant redesign. It is no surprise then that the successful contractors are discovering that their prized systems are needing to be rebuilt virtually from scratch.

The real sadness is that if those driving this scheme had thought differently, they could have harnessed the knowledge, expertise and enthusiasm behind the large numbers of successful small scale systems that were already in productive use throughout our NHS. As it is, a large cohort of “early adopters” now feels alienated and frustrated.

The more egotistical brand this as the The world’s biggest IT project. It may well be the largest, centrally-driven, top-down IT project on the Planet—but that is hardly a category with a conspicuous history of success. I suspect that the real contender for the “largest IT project” accolade must be the Internet, which is unquestionably hugely successful. Think for a moment about how the Internet came about: its project director, project manager, project plan, champions, whoever. Then read Tony Hoare’s 1980 Turing Lecture: The Emperor’s Old Clothes (especially his take on the classic Hans Andersen tale at the end, and the following quote).

At first I hoped that such a technically unsound project would collapse but I soon realized it was doomed to success. Almost anything in software can be implemented, sold, and even used given enough determination. There is nothing a mere scientist can say that will stand against the flood of a hundred million dollars. But there is one quality that cannot be purchased in this way—-and that is reliability. The price of reliability is the pursuit of the utmost simplicity. It is a price which the very rich find most hard to pay.

Then go place your bets.