sábado, setembro 30, 2006

Andreas Eschbach's Bibliography

For those of you who have shown an interest.

Novels

* Die Haarteppichknüpfer (1995), ISBN 3-7951-1371-7, translated as "The Carpet Makers" (2005), ISBN 0-7653-0593-3



* Solarstation (1996), ISBN 3-7951-1406-3





* Das Jesus Video (1998), ISBN 3-7951-1797-6






* Kelwitts Stern (1999), ISBN 3-7951-1624-4





* Quest (2001), ISBN 3-453-18773-3








* Eine Billion Dollar (2001), ISBN 3-7857-2049-1 ("One Trillion Dollars")






* Der Letzte seiner Art (2005), ISBN 3-404-15305-7














* Der Nobelpreis (2005), ISBN 3-7857-2219-2

Young Adult novels

* Perfect Copy: Die zweite Schöpfung (2002), ISBN 3-401-05425-2
* Das Marsprojekt (2004), ISBN 3-404-24332-3
* Die seltene Gabe (2004)

Short stories

* "Dolls"
* "Der Mann aus der Zukunft"
* "Die Wunder des Universums"
* "Humanic Park"
* "Halloween"
* "Eine unberührte Welt"
* "Unerlaubte Werbung"
* "Der Drache im Hindukusch"

Other

* Der Gesang der Stille (novella in the Perry Rhodan series)
* "Die Rückkehr" (his second Perry Rhodan novella as a "guest writer")
* Exponentialdrift (novel serialized in Frankfurter Allgemeine Zeitung)
* Eine Trillion Euro (short story anthology of European SF writers edited by
Eschbach)
* Das Buch von der Zukunft: Ein Reiseführer (2004), ISBN 3-87134-476-1, non-fiction

John Clute also casts a look upon the book "The Carpet Makers" ...

For anyone interested here goes his take upon the book published on his column "Excessive Candour":

http://www.scifi.com/sfw/issue417/excess.html

2005 Locus Recommended Reading List - "The Carpet Makers"

Quite true Pedro. I've just looked it up at home ( I've got the magazine at home ). It was published in the 2006 February issue. I'm not sure whether the list is available on their site. I can scan the list and post it here later on.

I've just ordered two more Eschbach's books : "Kelwitts Stern" ( ISBN : 3404232321 ) und "Solarstation" ( ISBN : 3404242599 ). "Solarstation" was the book after ...

Thanks for the tip.

sexta-feira, setembro 29, 2006

"The Carpet Makers" by Andreas Eschbach



I've been told that this book has already been published in english ! According to João the english translation is not clunky. Check it out at www.amazon.com.

"Die Haarteppichknüpfer" by Andreas Eschbach



I picked up this book purely by chance, but I was completely flabbergasted by it !

I've been reading Science Fiction for more than 25 years and I didn't expect to be blown away by this book. After all I think someone said that SF is dead and buried ! Not so ! This is not exactly Hard SF, but the scent is there. It reminds me of some of the Charles Stross' books ( serie "The Merchant Princes" ). Like Stross Eschbach is a true master of the form. It's not old-SF disguised as new. It's really new !

I haven't finished it yet, but I can't wait to write something about it. I've just gone on the net to find something more about this guy. It turns out that this was his first book, which came out in 1995 ! I think it should be translated into english so that more people can read it.

With his first work Eschbach shows that the German SF is not dead. The world sketched by it is just as extremely moving as strange, and without any spectacular action Supports or Super Aliens. By continuous jumps in character perspectives as well as Timelines, the author knows how to catch the reader's attention up to the end, until finally the multitude of mosaics result in a larger picture.
Even if the resolution and some other narrative aspects have more to do with Fantasy than with Science Fiction, it's still one hell of a book !

The structure of the book is very much like a collection of short-stories, but on a deeper level it's not. It's something different. There's always a thin thread that links all the chapters together ( like a tapestry ... ), but sometimes you've got to look closer, because it's hard to see and understand.

The first paragraph is quite extraordinary, because it sets the tone for the remainder of the book. It reads like a Fairy Tale, without being one ... :

"Knoten um Knoten, tagein, tagaus, ein Leben lang, immer die gleichen Handbewegungen, immer die gleichen Knoten in das feine Haar schlingend, so fein und winzig, daß die Finger zittrig wurden mit der Zeit und die Augen schwach von der Anstrengung des Sehens - und die Fortschritte waren kaum zu merken; wenn er gut vorankam, entstand in einem Tag ein neues Stück seines Teppichs, das vielleicht so groß war wie sein Fingernagel. So hockte er an dem knarrenden Knüpfrahmen, an dem schon sein Vater gesessen war und vor ihm dessen Vater, in der gleichen gebeugten Haltung, die alte, halbblinde Vergrößerungslinse vor den Augen, die Arme auf das abgewetzte Brustbrett gestützt und nur mit den Fingerspitzen die Knotennadel führend. So knüpfte er Knoten um Knoten in der seit Generationen überlieferten Weise, bis er in einen Trancezustand geriet, in dem ihm wohl war; sein Rücken hörte auf zu schmerzen, und er spürte das Alter nicht mehr, das ihm in den Knochen saß. Er lauschte auf die vielfältigen Geräusche des Hauses, das der Großvater seines Urgroßvaters erbaut hatte - den Wind, der ewig gleich über das Dach strich und sich in offenen Fenstern fing, das Klappern von Geschirr und die Gespräche seiner Frauen und Töchter unten in der Küche. Jedes Geräusch war ihm vertraut. Er hörte die Stimme der Weisen Frau heraus, die seit einigen Tagen im Haus lebte, weil Garliad, seine Nebenfrau, ihre Niederkunft erwartete. Er hörte die halbstumme Türglocke scheppern, dann ging die Haustür, und Aufregung kam in das Gemurmel der Gespräche. Das war wahrscheinlich die Händlerin, die heute kommen sollte mit Lebensmitteln, Stoffen und anderen Dingen."
( I won't even dream of translating this ... )

I hope the english edition will appear one of this days. Unfortunately I think much will be lost in the translation ( the poetry of the prose, the choosing of words, the interplay between sentences, etc ).

Those of you who know German go and a grab a copy. You won't regret it ! Those who don't, well ...

It's a crepuscular novel, filled with silences and things not said. Very strange and surreal indeed. But the strangest thing about this novel is that the protagonists of the Story, the Carpets themselves, are not described at all ! They are there, we are able to think about them, to imagine them. Maybe that's the all point.

Author's homepage : http://www.andreaseschbach.de/
( the abovementioned paragraph was extracted from this homepage )

quinta-feira, setembro 28, 2006

Witze auf Deutsch

http://www.niemiecki.ang.pl/Witz_9826.html

It's already late, so I'm not in the mood to translate !

Here they go :

"Der Anatomie-Professor zur Studentin: "Welcher Teil des menschlichen Körpers weitet sich bei Erregung um das Achtfache?" Sie wird rot und stottert: "Der..., das..." - "Falsch, die Pupille", entgegnet der Professor. "Und Ihnen, gnädiges Fräulein würde ich raten, mit nicht zu hohen Erwartungen in die Ehe zu gehen..."

-------------------------------------------------------------

Herbert trifft auf der Straße nach langer Zeit seinen alten Freund Gerd wieder: "Mann, dir sieht man ja auf 100 Meter schon an, dass du inzwischen verheiratet bist - frisch gewaschene Hemden, gebügelte Hosen, blanke Schuhe...!" - "Tja", sagt Gerd stolz, "das war auch das erste, was meine Frau mir beigebracht hat!"

-------------------------------------------------------------

Am frühen Morgen geht ein Mann auf die Jagd. Im Wald angekommen, beginnt es zu regnen, der Wind nimmt zu. Der Mann beschließt, umzukehren. Er kommt nach Hause, zieht sich aus und legt sich wieder zu seiner Frau ins Bett. "Wie ist es draußen?" fragt seine Frau gähnend im Halbschlaf. "Kalt, es regnet." - "Und mein Mann, der Idiot, ist auf die Jagd gegangen."

-------------------------------------------------------------

ein hartes Stück:
Kommt der Nikolaus in ein Behindertenheim und meint: "Wer mir ein Gedicht aufsagt, kriegt ein Geschenk. "Das 1. Kind sagt ein langes Gedicht auf. "Das war ein langes Gedicht und deshalb bekommst Du ein großes Geschenk." Das 2. Kind sagt ein kurzes Gedicht auf. "Das war ein kurzes Gedicht und deshalb bekommst Du ein kleines Geschenk." Meint das 3. Kind: "Nnng ek ninnnh dn drn!" - "Und wer mich verarschen will, kriegt gar nichts!

-------------------------------------------------------------

Ein Mann sitzt allein in einer Kneipe und muss auf die Toilette. Er hat allerdings Angst, dass ein anderen sein Bier trinkt, wenn er weggeht. Also hängt er eine Zettel dran: "Ich habe reingespuckt!" Als er wieder zurückkommt, steht darunter "Ich auch."

-------------------------------------------------------------

Morgens, 7 Uhr. Die Ehefrau stellt dem Beamten das Frühstück vor die Nase, inklusive Zeitung. Sie essen, er liest die Zeitung, keiner sagt etwas. Drei Stunden später sitzt er immer noch am Tisch, liest die Zeitung, nickt ab und zu ein, schaut manchmal aus dem Fenster... Da sagt die Frau: "Sag mal, Schatz, musst du heute gar nicht ins Büro fahren?" Er springt erschrocken auf: "Mist, ich dachte, ich wäre längst da..."

-------------------------------------------------------------

PMP Question - Monte Carlo Analysis and DOE

Rita sent me an interesting question. Because it's a hard one to crack I'll let you all try to answer it before I do...

Send your contributions to the usual email account.

"Could anyone please compare/contrast Monte-Carlo analysis and Design of Experiments?"

I'll post the best answer ... ;)

PMP Question - Probability of Project Schedule

"Manuel, can you tell me the answer for the following question and the reason behind this ?

If you have a 15% probability of exceeding the project schedule you are
a. Above the mean
b. Below the mean
c. At mean"

Another question with some possible different interpretations.

Answer :

The probability that we cannot keep the project schedule is 15%. Duration of our project schedule is:

A. Above the mean of estimates.

"Exceeding the project schedule” surely means that the actual finish date would be behind the determined schedule. And if the probability of delay (so-called failure) is 15%, the probability of success is 85% (100-15).

The farther the determined schedule is from the mean (ABOVE the mean), the less the probability of delay becomes.

This a clear example of inversed reasoning ... LOL. In terms of statistics this is very common.

PMP Question - Standard Deviation

Matthias, thanks for the question ( transcribed below ).

Question :

"I have a problem with one questin about time management.
The question is:
The project manager for a project is going to use the PERT method of scheduling for the project that she is working on. using PERT calculation the variance fr the project is found to be 25 days. What is the range of values for the project duration such that here will be a least 95% probability that the actual project completion will fall between the high and low value of range of values:
a. 164 - 173 days
b. 144 - 194 days
c. 119 - 219 days
d. 159 - 179 days

My answer:

The point of this question was to recognize that the variance is the standard deviation squared. For a variance of 25 the standard deviation should be 5. 2 sd is 10 and +- 10 is the range of values that is 95% probability of including the actual value. The range of values is +-10 or 20. Answer "A" is 9, answer "B" is 50, answer "C" is 100 and answer "D" is 20. D is the right answer.
The answer in book is D but from my calculations score is 152,4 -
185,6 days. Could you give me a proper answer. Maybe I did something
wrong.
My calcutations do not fit to any point."

Answer :

Variance=(standard deviation)squared so sd=5
that means that the range of values is from -5 to +5

95% probability means two times sd.
that means the range is -10 to +10 i.e. 20

This means that "D" is the correct answer !

Simple ah !

PMP Question - Privity of contract

That's a mouthful !!

Those of you who are privy ( sic ) to this sort of thing, may know how to utter this in portuguese but not me ( can anyone help ? "Prividade" ??? )!

In german it's easier because one just have to concatenate two words : "contract" ( "Vertrag" ) and "Relation / Relationship" ( "Beziehung" ) and voilá we have the new word : "Vertragsbeziehung", which means literaly "Contract Relationship".

In english "Privity" is just a fancy word for relationship ! In down-to-earth language is just plain used when two companies have a contractual relationship with each other.

Having dealt with this let's turn to the question in hand ( sent by Wolfgang ):

"When two companies have privity, what happens if one of those companies hires another company to do some work for them ? The Privity between the first two companies extends to the third company ?"

The answer is no. As we have seen above, "Privity" means relationship, and there isn't one between the first company and the third company. That means that the Commutative Law does not apply ... LOL.

Let's try to see this with an example. Company A has privity ( CONTRACTUAL RELATIONSHIP ) with Company B, but company B doesn't have the manpower or the skills to tackle this project on its own, so it hires company C. A more difficult question to pose is : "Has the Project Manager of company A a mandadory influence on what the Project Manager of company C does ?". That is the question. The answer is once again no, due to the fact that there's no contractual relationship ( Privity ) between company A and company C ! That means that company B must be involved at all times.

quarta-feira, setembro 27, 2006

PMP Question - Portfolio Management

Is Portfolio Management ( PM ) the same as the Business Case ( BC ) ? That is the question ...

In terms of Project Management this takes place in the process of Develop Project Charter by using the tool Project Selection. Before developing the Project Charter we've got to make sure that the project in question is aligned with the IT strategy of the company. PM addresses this issues directly ( among other things ).

It goes hand-in-hand with the Business Case ( BC ). I usually think that Portfolio Management comprises the BC.

The several methods that comprise the Project Selection ( ROI, NPV, BCR, etc ), all of them can be used when setting up a BC. The concept of PM is more encompassing, that is, we're deeply concerned with IT Alignment, Kill project criteria, and so forth. The bottom line is that by using PM we are accepting the fact the current criteria for Project Selection may be non-existent. By establishing Project Selection criteria we're in fact saying that we've got a Framework for choosing projects. Why use BCR and not ROI or IRR ? Should these criteria be used in tandem ? What is the underlying framework that consistently allows us to choose between projects ? Should the criteria be changing all the time ? Those are the questions that Portfolio Management tries to address. By saying this, it is straightforward that the BC is a subset of the Portfolio Management, ie, one should not be confused with the other ( they're not interchangeable ).

Portfolio Management is the selection and support of project or project investments as guided by the organization's strategic plan and available resources. That means we've got to be damn sure that we're choosing the right projects and we're also are in possession of the tools to back up those selections / decisions !

sexta-feira, setembro 08, 2006

PMP Question - IRR ( TIR in portuguese ) in Project Analysis

Question :

"The Internal Rate of Return formula as a method of project analysis:

a. uses the opportunity cost of funds as the discount rate in its formula.

b. solves for the interest rate that makes the net present value of benefits and costs from a project equal to zero.

c. results in ambiguity in some cases since there will generally be more than one internal rate of return solving its formula.

d. is preferred to net present value for business projects but not for environmental projects.

e. both b. and c. are true."

Answer :

To analyse this question we must first address the question on how to calculate the IRR.

If the internal rate of return is less than the cost of borrowing used to fund your project, the project will clearly be a money-loser. However, usually a business owner will insist that in order to be acceptable, a project must be expected to earn an IRR that is at least several percentage points higher than the cost of borrowing, to compensate the company for its risk, time, and trouble associated with the project.
IRR analysis is generally used to evaluate the project's cash flows ( CF ). That means we need to use the NPV formula in order to get the IRR :



If we make NPV=0, given the CFs, we have a polynomial with n-root. As lots of you will remember, it's impossible to solve this equation analitically. That means we must use an iterative method in order to find the root ( r = IRR ) of the polynomial.
By making NPV=0, we are in fact saying that IRR is the highest rate by which someone might borrow money for an investment without loosing money !

By looking at the options it's easy to see that "b" must be true. Due to the fact that we are trying to solve a polynomial, it means that we might get more than one solution to the equation ! That means that "c" must also be true.

All in all the correct answer is "e" !

terça-feira, setembro 05, 2006

PMP Question ? - Probability Distributions

I'm not sure this is really necessary for the PMP Exam, but better safe than sorry.

I had to look it up in the book "Introduction to the Theory and Practice of Econometrics" by George G. Judge et al ( ISBN = 0-471-60272-8 ). I've got the 2nd edition of this book. Anyone interested in buying it should get the last edition.

Here it goes.

Binomial Distribution (Success or Failure)

• A coin will be tossed 5 times but the coin is biased so that the probability of heads for each toss is 0.04. Heads is success, tails is failure.
• N = number of items in the sample (the number of coin tosses)
• X = number of items for which the probability is desired (number of Heads)
• In Appendix A we go to column N and find where N = 5
• In Appendix A we go to where p = 0.40
• Each row represents the probability of 0, 1,2, 3, 4, and 5 successes
• Add them up

Poisson Distribution

• A lightbulb manufacturer has a known defective rate of 4%. From a sample of 40, the probability of 4 or more defective light
• µ = np = (40) (.04) = 1.6
• Probability of 4 or more defective is = 1 – probability of 3 or less defective
• In the table, find where µ = 1.6
• Add up the numbers where x has a value of 0, 1, 2, or 3 (this is the P of 3 or less defectives)
• Subtract that number from 1.0
• Find np (sample x defective rate)
• Calculate up to by going to the table, finding np, adding it up
• Subtract that answer from 1 to x or greater probability

Normal Distribution (also known as Gaussian)

• If process produces parts with mean of _ and standard Deviation of _, what is the P that one random part has a measurement of _?
• Mean time of a bank transaction is 5.25 with a standard deviation of 0.75 minutes and the values are normally distributed. What is the probability that a transaction will occur between 4.0 and 5.25 minutes and below 4.0?
• Z = 4.0 – 5.25/ 0.75 = -1.67
• Go to Appendix A and find 1.67 = 0.4525
• Because we know that µ is 5.25, the probability that a transaction will take less than 5.25 is .05 (1/2)
• Therefore, the probability that a transaction will be less than 4 minutes = 0.5 – 0.4525 = 0.0475

Sampling Distributions (number of standard Deviations that a sample mean is away from the population mean)

• If normal distribution with mean of _ and SD of _. From sample of _ what is P that the sample mean is >, <, =, or between _?
• Hospital emergency room where it has a record waiting time of 30 minutes with a standard deviation of 5 minutes. If a sample of 35 is measured, what is the probability that the sample mean would be greater than 31.5 minutes?
• Do the Z calculation to get 1.77
• Find 1.77 in Appendix A (go to 1.7 and then across to 0.07)
• Subtract that probability from the .5 probability = .50 - .4616 = .0384
• This tells us that there is only a .0384 probability that, from the sample of 35, the sample mean will be greater than 31.5.

I hope it will help ...

NB : The appendix A referred in the notes contains all the tables that allow the calculation of the probabilities. On this day and age, I always use the scientific calculator ...

PMP - Formula Derivation : ( N ( N - 1 ) )/2

Wolfgang thanks for the question ... LOL ! This one was hard to crack ! I had to make a few assumptions to get it right. I would like to have your feedback on this one, because the derivation is tricky.

I hope the explanation will be crystal clear.

This calculation will help to realize the concept of communication channel.
Say, there are “n” Persons and they are having communication with each other. We need to find how many communication channels are possible here.
Take the case of the first person. He can communicate with rest of the (n-1) person. So, here are (n-1) communication channels.
The Second person can communicate with ((n-1)-1) person because his communication with the first person is already taken into account for the calculation of first person. So, here are (n-2) communication channels.
Similarly, the third person can communicate with ((n-1)-2) because his communication with the first and second person is already taken into account for the calculation of first and second person. So, it’s (n-3) channels.
Like that, if we proceed, the total number of Channels used by n person, will be = (n-1)+(n-2)+(n-3)+......+1 = 1+2+3+.....+(n-1)

We have a sum of a series for "(n-1)" number whose first number is "1" and the common difference is also "1".
The Formula for Sum of a series is: S = n[2a+(n-1)d]/2
(where, n = total numbers, a = first number, d = common difference)

So, it gives us the Total Communication Channel
=(n-1)*[2*1+((n-1)-1)*1]/2 = (n-1)*[2+(n-2)]/2 = n*(n-1)/2

PMP Question - Probability that a process will finish in less than x days

( Wolfgang, I think you should read from scratch the chapter on Time Management of the PMBoK and the same chapter of the book "Das PMP-Examen" ).

Question :

"A process has Critical path of 12 days & Standard Deviation of 1. What is the probability that the process will be completed in 13 days.
A : 10%
B : 68%
C : 84%
D : 95%"

Answer: C ! I bet you would have chosen B. Wrong !! ( As the target date is 1 SD away, one might think that the right answer would be 68% ...).

The probability of PERT estimates is 50% and it is the mean.
1 SD from the mean covers an additional population of 34.1%, 2 SD from the mean covers 47.7% and 3 SD covers 49.8%. Conversely, -1 SD from the mean covers a population of -34.1% etc.
13 days means - 1 additional day from the mean (that is 1 SD too); so from the data above the cumulative probability to complete, comes from 50% + 34.1% = 84.1% (answer C).

To better understand my affirmation, we should consider a PERT estimate as a mean of a normal distribution (although PERT is based on a beta distribution but we all know that all the distribution can be approximated by a normal distribution), thus the area under the Gaussian curve delimited by the mean is 50% of the entire area.

As with many PMP math questions, one has to read between the lines, to perceive what is really being asked.

PMP Question - Standard Error

Question :

"A process characteristic has a true mean of 150 and a true standard deviation of 20. A sample of 25 items is taken from this process. The standard error of the estimate is:
A. 0.2
B. 0.8
C. 4
D. 5
E. 7.5"

Correct Answer is C.

The Standard Error of a statistic is the “Standard Deviation” of the “sampling distribution” of that statistic. Standard errors are important because they reflect how much sampling fluctuation a statistic will show. The standard error of a statistic depends on the sample size. In general, the larger the sample size, the smaller the standard error. The standard error of a statistic is usually designated by the Greek letter sigma (sigma) with a subscript indicating the statistic. The formula for the standard error of the mean is:



where sigma ( numerator ) is the standard deviation of the original distribution and N is the sample size (the number of scores each mean is based upon).

So, Standard Error here is 20/sqrt(25) = 20/5 = 4. That is C !

The catch here was to know that a "standard error" is the same as "standard deviation". Tricky ...

PMP Question - Task Duration

Question :

"An Activity has an Optimistic estimate of 10 days, pessimistic estimate of 16 days, most likely estimate of 13 days. If your company has a quality requirement of 6 sigma, what is the duration within which this task must be completed?

A. 10 days to 16 days
B. 7 days to 19 days
C. 12 days to 14 days
D. 11 days to 15 days"

Answer :

In a Normal Distribution, the PERT duration (also called mean)
= (Pessimistic + 4* (Most Likely) + Optimistic)/6
= (10 + 4 * 13 + 16)/6
= 13

1 Standard Deviation (sigma)
= (Pessimistic - Optimistic) / 6
= (16 - 10) / 6
= 1
Mean - 6 sigma = 13 - 6* 1 = 7
Mean + 6 sigma = 13 + 6* 1 = 19

So, if the the company has a quality requirement of 6 sigma, then the task can be completed within 7 days to 19 days ( option B ).

PMP Question - Difference between Beta distribution and Triangular Distribution



The statistical distributions are used based on the type of situation you encounter during execution of the projects.

Pert Methodology of estimation of miu =(to+4tm+tp)/6 considers a Normal distribution. Where you have large number of samples (in other words Historical Information) available.

A Triangular distribution can be preferred when you have estimates based on the judgemental data submitted to you.

In all the cases what is more important from Project Management Perspective is the kind of Probability they carry for overall estimate of the project.

While Pert Method will have a smooth distribution the Triangular will have an abrupt peak at miu.

I would prefer to use Pert Method when I am having an environment of similar works being done earlier, this gives a larger spread of variability. And Triangular for an Extremely New Environment. And minimize my risks of Schedule overruns. In a small variability of the Overall Schdule.

Formulas:


If distribution is triangular , then mean is :



If the distribution is beta, then the mean is :