The Great Neo-Hapsburg Euro-Apagão of 2006

Cool Hand Luke and risk management: “What we have here, is a failure, to commun’cate.”

The latest from the Risks Digest — from the ACM FORUM ON RISKS TO THE PUBLIC IN COMPUTERS AND RELATED SYSTEMS — includes this assessment of interjurisdictional electricity grid mismanagement by those normally punctual, efficient, precise and humorless Germans.

The ACM is one of those organizations I pay dues to, by the way, because I support what they do — and, just as important, how they do it. So far, at least.

Like the ACLU, the NWU, and the SABEW.

Unlike, very pointedly, the SPJ and the NYFWA, with special honors to the maracutaia artists at the ASBPE.

I have been thinking I should do a write-up of the air traffic control mini- or midsize crisis — creeping towards macro if it continues on past Christmas — currently backing up airports here in Brazil, and submit it to this top-notch meeting of the minds.

I did once have a note published there, you know. I am proud of that.

But frankly, those MIT risk geeks, as nice as they are personally, intimidate me. I am afraid of making an ass of myself, believe it or not.

But you never learn things unless you risk just that. What I always tell my students, when I am in teaching mode.

Anyway, let’s annotate the RISKS-24.28 on the Great Euro-Apagão of 2006.

As always, circle words you do not know for later consultation with a dictionary.

More details have emerged on the EON Austria-to-Spain power outage since in RISKS-24.47 (which erroneously stated rather absurdly that 82 million Germans were affected, instead of the previously noted 10 million Europeans).

Axel Eble cited the original German text of the E.ON Netz report:

[URL]

Jaap Akkerhuis cited the E.ON report in English:

[ URL]

The upshot is that the initial calculations for the planned shutdown showed the link over the Ems River could be compensated for by rerouting alternative power. The so-called “N-1 criterion” for stability was correctly applied initially, but not reapplied after the reconfiguration.

Thus, the second-order effects of the shutdown were ignored — namely the increased loads that would result from the rerouting — and the Norwegian Pearl was allowed to pass.

From the English language version of the report (which explicitly notes that the German version shall prevail in case of any discrepancies), the summary states that “the determination of demands that can actually be met and which the market participants demand of the grid must be continuously be reviewed in a close dialogue between grid operators, grid customers, regulating authorities and political forces.” [*]

Continuing from the summary, “Finally, it also remains to be stated that the concrete incident has no connection with issues of grid investments. It must, however, be clearly stated that the growing demands on the grid can only be met — in the long run — by a corresponding expansion of the grids.”

Once again, we are confronted with the risks of short-term/global
optimization.

Aka “the tragedy of the last mile.”

Bonus track: The NMM “I am not the only one who digresses like crazy” award for the apparently irrelevant but actually on-point aside of the week.

[* NOTE: As a rather PGN-ish aside, Webster maintains that a "dialogue" can be BETWEEN N entities, where N may be two or more. However, when I learned English, it was customary to make a distinction between "between" and "among" (for N=2 and N>2, respectively).

Someone taught you well. I am a nasty old-school stickler on that point myself. --Ed.

This seems to have fallen by the wayside over time. On the other hand, German uses one word ("zwischen") to cover both, as do French/Spanish ("entre") and Russian ("myeshdu"). At any rate, the concept of a CLOSE DIALOGUE BETWEEN (or even AMONG) N entities seems suspect when N is considerably greater than 2, as it is in the European community, and when communication is inherently NOT CLOSE. I think that the choice of the German text ("im engen Dialog von ...") is itself misleading, and that the English translation could have been more accurately rendered as "in close multipartite communication among ...".

Yes, dialogue comes from a Greek word meaning "a conversation between two parties" and has as its primary scene-act ratio the branch of rhetoric known as forensic, which prepares the citizen in training to practice effectively in the two-sided adversarial system of the law courts.

Multiparty, multitopic discussions -- which we do sometimes refer to, informally but oxymoronially, as "dialogue," but which formally speaking we classify as "debate," "deliberation" and perhaps you could throw "negotiation" in there in some cases -- belong to the deliberative branch, the most advanced form you learn in the traditional humanist curriculum of the citizen in training, for which the previous stages are understood as preparing the way.

Its purpose is to prepare the student for functioning effectively in the democratic assembly -- the fullest degree of participation in the Greek polis and Roman res publica, to which all citizens are expected to aspire.

Technology "evangelism," by the way, belongs to the epideictic branch -- the lowest rung on the ladder, the easiest to learn and abuse, and whose effects are the cheapest to produce ... though it is not without its pedgagogical and commercial interest.

Its typical zone of applicability is the agora, or marketplace.

It is especially useful for addressing the gullible and the ignorant, in order to save oneself the time and effort required to close the deal through a negotiation inter pares in which no cost-efficient information arbitrage opportunities are available to either side. -- Ed.

Why do I engage in such semantic blather? Because the lack of CLOSER COMMUNICATION is often a serious source of risks in many RISKS episodes, and Conway's Law and generalization thereof keep resurfacing as representative of fundamental problems that arise from restricted communications.

(Wikipedia has a nice discussion of Conway's Law, which relates difficulties in communication specifically to corresponding flaws in software developments. However, certainly someone must have cited its obvious generalization to other types of systems.

Surprisingly, I don't think I've mentioned Conway's Law previously in RISKS, although I have been referring to it explicitly and in its generalized forms for many years. Melvin, not John.) PGN]

I’m glad to know that I did not completely waste my time reading the works of pre-Habermasian semanticist and U.S. Sen. S.I. Hayakawa (Dem.-HI) in college.

There is an elaborate Brazilian shaggy dog storyfloating around the folkloric Tabajara-sphere down here that I can never remember properly, but that boils down to something like this:

In heaven, all the engineers are German and all the musicians are Brazilian. In hell, it’s the other way around.

But take solace, my Tupi companheiros — and oompah tubists, too, for I was once one of you as well.

As this case shows, mito é mito e Deus é Deus.

About these ads