Nuunnnnn wal qolami wamaa yasturuun.....

Kepada rekan-rekan yang mempunyai artikel atau informasi apa saja untuk blog ini silahkan kirimkan ke weld_inspect@yahoo.com / mochpahlevi@yahoo.com

sementara blog ini masih dalam proses penyempurnaan, untuk itu kepada rekan-rekan yang mau memberikan ide, saran dan kritiknya silahkan kirim email ke weld_inspect@yahoo.com / mochpahlevi@yahoo.co.

kami ucapkan jazaakumullah khairon katsiiro atas perhatiannya

wassalam

mein Name ist Simon Krebs und ich bin Existenzgründer. Ich möchte Ihnen auf diesem Weg meine Geschichte und die Geschichte meines Unternehmens erzählen. Natürlich ist es immer schwierig, für eine gute Geschichte auch einen guten Einstieg zu finden, vor allem, wenn einem praktisch jeder im Internet dabei zusehen kann. Ich werden den Anfang der Geschichte da ansetzen, als ich die Idee für mein Geschäft hatte - der Ausgangspunkt:

Pizzaporter.de

Dieses Kapitel widme ich allen Leuten in meinem Familien-, Freundes- und Bekannten-Kreis, die sich an die Phase in meinem Leben noch gut erinnern! Es war Anfang des Jahres 2006; damals hatte ich eine - in meinen Augen - ziemlich faszinierende Idee: Ich würde mir die fortschreitende Kommerzialisierung, die auch schon einige konservativ geprägte Branchen erfasst hatte zunutze machen.

• Amazon hatte es geschafft, Leute dazu zu bewegen, Bücher im Internet zu bestellen.
• Ebay hatte es geschafft, Leute dazu zu bewegen, Flohmärkte im Internet abzuhalten.
• Docmorris hatte die Idee, sich mit der konservativsten aller Sparten - Apotheken - einen Wettkampf um die Gunst der Kunden zu liefern.
• ... und ich hatte die Idee, Menschen über das Internet mit Fast-Food zu beliefern. Konkret: und das sagt auch schon der Name, ging es um die Bestellung von Pizza!

Warum nicht? Wir alle kennen die Probleme: Sie müssen das Wort S-a-l-a-m-i buchstabieren, damit Ihr Gegenüber am Telefon versteht, was Sie meinen. U.U. kommt die heiß ersehnte Mahlzeit nie an, da der Pizzamann die Adresse falsch verstanden hat. Und bevor Sie überhaupt etwas bestellen können, müssen Sie zuerst einmal eine Telefonnummer finden (lassen Sie mich kurz überlegen: Letzte Woche habe ich keinen Flyer von einem Pizzaservice bekommen, was praktisch bedeutet, dass ich keine Nummer habe!). Im Zweifelsfall hilft nur eines: Das Internet befragen!

So - und wer schon soweit gekommen ist, der wünscht sich vielleicht eine Möglichkeit, die Pizza direkt bestellen zu können. Tja, die Idee war gut - davon bin ich heute noch überzeugt. Mit viel Energie habe ich mich an die Arbeit gemacht und recherchiert: Gibt es sowas schon? - Nichts ernstzunehmendes, also ran an die Arbeit:

Businessplan geschrieben, Plattform programmiert + gehostet, weiterprogrammiert, optimiert und schließlich auf die Suche nach einem Investor gemacht. Denn eine Sache ist sicher: Ebay, Amazon und Co. wären heute nichts, wenn Sie damals nicht schon ein hervorragendes Marketing gehabt hätten. Und wie schnell man zur Nr. 2 wird hat man damals an Ricardo gesehen, die ja lange Zeit stark mit Ebay konkurriert haben. Die Gründungsvorbereitung ist ziemlich gut verlaufen. Pizzaporter hat sich in der Entwicklungszeit zu einer ausgewachsenen Plattform etabliert mit sämtlichen Schnick-Schnack, den ein modernes E-Commerce so benötigt. Community-Elemente, ein Wiki und vor allem ein gutes Backend und eine Infrastruktur, die mehrere hundert Bestellungen am Tag vollautomatisch und mit Sicherheits- und Rückkopplungs-Mechanismen verarbeiten könnte. Auch die ersten Lieferanten wurden an das System geklemmt, nachdem wir Sie in mühevoller Arbeit von dem System Pizzaporter überzeugen konnten. Und auch der Businessplan war Mitte 2006 soweit, dass wir damit auf den Kapitalmarkt gehen konnten!

Die Sache hatte nur einen Haken: Der Markt war zu dem Zeitpunkt, als ich Ihn mit Pizzaporter betreten wollte, schon in der Konsolidierungsphase. Ich hatte bei meiner Marktanalyse einen Wettbewerber unterschätzt: bringdienst.de hatte, während ich mit der Entwicklung der Plattform beschäftigt war, fleißig damit weitergemacht, die Großen der Branche - Joeys Pizza, Pizza Dino, Hallo Pizza, und wie Sie alle heißen - als Kooperationspartner geworben, sprich: jede Bestellung, die online bei den großen Ketten eingeht, wird über die Infrastruktur von bringdienst.de abgewickelt. Und die Konkurrenz hatte noch einen zweiten Schlag geplant: Im Herbst 2006 übernahm bringdienst.de die Plattform und die Internetadresse pizza.de, brachte einen Relaunch der eigenen Plattform heraus (inkl. all der schönen Schnick-Schnack-Elemente, die ich oben erwähnt habe) und fing an, Werbung für die eigene Sache zu machen. Vorher hatte bringdienst.de einige Lücken im Leistungsprogramm, die ich mir durch eine clevere Positionierung zunutze machen wollte. In meinem Businessplan heißt es zu dem Konkurrenten:

bringdienst.de

Stärken:

• flächendeckender Lieferantenstamm
• Bestellabwicklung über Internetauftritte von Lieferanten
• Zuverlässigkeit der Lieferanten (Franchise-Ketten)

Schwächen:

• unattraktiver Internetauftritt
• weder Content-, noch Community-Elemente auf der eigenen Plattform

Mit den oben beschrieben drei Schritten hob mir bringdienst.de förmlich alle USPs meiner Geschäftsidee aus den Angeln - und das, kurz bevor ich selbst mit meiner Plattform starten wollte!

Nun ja: man könnte sagen - dumm gelaufen! Aber wenn ich die Sache jetzt ehrlich im Rückblick betrachte kann ich sagen: Ich habe jede Menge bei dem Projekt gelernt, unendlich viel Zeit dabei verbrannt (ich war quasi für das gesamte 3/4 Jahr für die Öffentlichkeit nicht mehr existent!) und Gott-sei-Dank rechtzeitig die Notbremse gezogen. Die Entscheidung, vor dem großen Konkurrenten zu kapitulieren, war schwer und wurde sehr bewusst getroffen. Trotzdem bin ich davon überzeugt, dass es richtig war. Hätte ich die Idee 1 Jahr vorher gehabt, so wären die Karten anders gemischt gewesen. So jedoch hätte ich mich gegen einen Konkurrenten mit enormen Vorsprung in einen Marketing-Krieg begeben müssen, aus dem mit Sicherheit nur ein Unternehmen als Sieger hervorgehen kann. Man nehmen sich die o.g. Beispiele Ebay und Amazon - wieviele Konkurrenten können Sie adhoc nennen ohne zu Überlegen?

Und dann?

Nachdem das Kapitel Pizzaporter abgehakt war, hatte ich zunächst einmal damit zu kämpfen, wieder in einen normalen Alltag hineinzufinden. Meine Noten während des Studiums sind in der Zeit rapide schlechter geworden, meine Anwesen- und Aufmerksamkeit während der Vorlesungen war ziemlich miserabel und der Kontakt zu einigen meiner Freunde fast vollkommen zum Erliegen gekommen. Heute verzeihen Sie es mir Gott-sei-Dank alle!

Ja, das Unternehmertum liegt einem aber entweder im Blut, oder nicht. Und bei mir tut es das definitiv! Deswegen habe ich kurz darauf wieder angefangen, nach einer neuen Idee zu suchen. Diesmal sollte es nur etwas bodenständiger zugehen. Trotzdem: der Fokus "Internet" war geblieben, denn das ist ein Thema, das mich in allen Facetten seit dem ersten Kontakt (das war 1998 mit einem Modemanschluss von VR-Web ;-)) schon immer sehr fasziniert hat. Es gibt schließlich keine Innovation, die ich auch bewusst erlebt habe, die unser Leben so extrem verändert hat!

Und so hat es nicht lange gedauert und ich habe von neuem angefangen, einen Businessplan zu schreiben. Wie es dazu kam, was genau ich jetzt mache und wie sich das alles entwickelt hat, das werde ich berichten - aber nicht mehr heute!

Bis dahin,

Simon Krebs

P.S. Trotz allem ist es schön und befriedigend, wenn immer mal wieder ein Pizzabäcker anruft und in unverständlichem, gebrochen Deutsch darum bittet, auf meiner Plattform gelistet zu werden. Das tue ich allerdings nicht mehr; die Plattform steht mehr oder weniger nur noch als ungenutztes Relikt im Netz!

So by effectively cutting sleep out of my schedule, I have managed to make colours appear in the magical land of the 'editor'.  Through some simple Abstract Syntax Tree crawling, I can easily identify all of the nodes I want to highlight and their offsets.  Finally, the benefits of writing my own AST and Parser are shining through and man does it feel good.  In any case, the first step will to get everything coloured in some sort of rainbowish attrocity and then to move everything to the preference pages.

However, the hurdle of getting things coloured has presented a few new issues:

• Resiliency; when the code isn't well-formed... how do we deal with it?
• Concurrency issues
• Parser problems (I knew it wasn't perfect, it couldn't have been that easy...)

Answers to all of these problems will come soon!  Until then, I must sleep!

[RO]"Prolog din proloage. Lecturi patristice dintr-un calendar uitat" de Preot Sever Negrescu.

[ENG]"Prologue. Patristic readings from a forgotten time" (huh?! nu stiu cum sa traduc... chiar nu inteleg fraza  :) )

:) "Ambiguity" is the word.

For Prolog [the programming language], click here .

tiba-tiba saja saya tergerak untuk membuat blog tentang psikologi. Sebenarnya sudah banyak blog yang sudah saya buat, haha namun semua berbicara masalah-masalah keseharian dan bagaimana saya memandangnya.

Dua minggu ini saya memulai masa stage, yaitu praktek sebagai calon psikolog klinis di salah satu rumah sakit Jiwa. Selama waktu yang singkat ini saya mendapatkan banyak sekali pengalaman baru yang sayang jika tidak saya tuliskan sehingga orang-orang lain dapat membacanya. Saya sadar betul bahwa banyak psikolog maupun mahasiswa sekalipun yang mempunyai pengalaman dan pengetahuan yang jauh melebihi apa yang saya miliki sekarang. Saya pun tau bahwa sedikit dari mereka yang punya cukup energi atau waktu untuk membaginya kepada masyarakat luas. Untuk memasyarakatkan psikologi sehingga menjadi ilmu yang "biasa" ada diantara masyarakat.

Saat ini, seringkali saya merasa, banyak permasalahan yang dikeluhkan oleh saya, rekan sesama mahasiswa atau senior saya yang telah menjadi psikolog. Permasalahan-permasalahan seperti bocornya alat-alat ukur psikotes, banyaknya orang-orang "tidak berkompeten" untuk melakukan pengetesan maupun intepretasi, orang "gila" yang hanya diberi obat tanpa diberi bimbingan intensif secara psikologis untuk membenahi hidup mereka, psikiater yang beranggapan bahwa pasien yang dapat "dikendalikan" sudah boleh pulang, masyarakat yang masih memandang miring orang-orang yang masuk RSJ atau yang berkunjung ke psikolog, dan lain sebagainya.

Sampai saat ini saya dan teman-teman hanya banyak protes dan mengeluh. Tapi menurut saya, memang sudah saatnya psikologi "dimasyarakatkan", jadi bukan hanya psikolog atau mahasiswa psikologi yang pintar urusan kepribadian manusia. Sehingga masyarakat luas pun sadar bahwa kode etik psikologi dibuat bukan hanya sebatas melindungi "lahan mencari uang" para psikolog namun sebenarnya juga untuk memastikan kesejahteraan klien/pasien terjamin.

Terlalu banyak untuk sebuah prolog... Dasarnya saya adalah seseorang yang memilih untuk mempelajari psikologi dan memutuskan untuk memasukkan psikologi sebagai bagian hidup saya. Melihat keadaan lingkungan saya,saya ingin membagi sedikit pengetahuan saya kepada orang yang berniat untuk membacanya. Terkadang saya memiliki beberapa pertanyaan, atau protes yang mungkin hasil pemikiran saya yang belum matang, namun untuk itulah saya ingin adanya suatu diskusi.

Apapun jadinya blog ini... pada akhirnya saya senang bahwa saya telah berusaha mengungkapkan apa yang saya pikirkan dan berusaha membantu sedikit demi sedikit.

Here's a more friendly guide to getting this awesome code to work. Skip to "code samples" below for awesome examples of this embedded language in action.

My system setup

I am running Ubuntu Hardy with SWI-Prolog 5.6.47; I installed SWI-Prolog straight from the Debian packages (through Synaptic, the usual way). Just get yourself a copy of Ubuntu and install SWI Prolog the sane way and this code should work.

The interpreter

change_member(_, N, [], [N]).
change_member(O, N, [O|XS], [N|XS]).
change_member(O, N, [X|XS], [X|YS]) :- change_member(O, N, XS, YS).

:- op(900, xfy, :=).
:- op(100, fx, [if]).
:- op(125, fx, [while,  return]).
:- op(125, yfx, [do, then, else]).

run({Block},                          EnvI, O, Ret) :-
  run(Block, EnvI, O, Ret).

run((First; Second),                  EnvI, O, Ret) :-
  run(First, EnvI, M, _), run(Second, M, O, Ret).

run(Place := Expr,                    EnvI, O, void) :-
  eval(Expr, Value, EnvI), change_member(Place = _, Place = Value, EnvI, O), !.

run(if {Cond} then {This} else {That}, EnvI, O, Ret) :-
  eval(Cond, Value, EnvI),
  ( Value = true -> Body = This ; Body = That ),
  run(Body, EnvI, O, Ret).

run(while(Cond) do {Body},            EnvI, O, Ret) :-
  eval(Cond, Value, EnvI),
  ( Value = true -> run(Body, EnvI, M, _),
                    run(while(Cond) do {Body}, M, O, Ret)
  ; O = EnvI, Ret = void
  ).

run(return Value,                     EnvI, O, Ret) :-
  eval(Value, Ret, EnvI), O = EnvI.

eval(Var,   Val,   Env) :- member(Var = Val, Env), !.
eval(Num,   Num,   _  ) :- number(Num), !.
eval(true,  true,  _  ).
eval(false, false, _  ).

eval(X + Y, Z, Env) :- eval(X, XV, Env), eval(Y, YV, Env), Z is XV + YV.
eval(X - Y, Z, Env) :- eval(X, XV, Env), eval(Y, YV, Env), Z is XV - YV.
eval(X * Y, Z, Env) :- eval(X, XV, Env), eval(Y, YV, Env), Z is XV * YV.
eval(X / Y, Z, Env) :- eval(X, XV, Env), eval(Y, YV, Env), Z is truncate(XV / YV).

eval(X = Y, B, Env) :- eval(X, XV, Env), eval(Y, YV, Env),
  ( XV = YV -> B = true ; B = false ).
eval(X < Y, B, Env) :- eval(X, XV, Env), eval(Y, YV, Env),
  ( XV < YV -> B = true ; B = false ).
eval(X > Y, B, Env) :- eval(X, XV, Env), eval(Y, YV, Env),
  ( XV > YV -> B = true ; B = false ).
eval(not(P), B, Env) :- eval(P, PV, Env),
  ( PV = true -> B = false ; B = true ).

Language samples

Here are samples of using this little embedded language we've built along with the evaluation results in comments.

factorial(N, {i := 1;
              n := N;
              while(n > 0) do {
                i := i * n;
                n := n - 1
              };
              return i}).

%% ?- factorial(5, P), run(P, [], _, X).

%% X = 120

fibonacci(F, {i := 1; a := 1; b := 1;
              while(not(i = F)) do {
                a := a + b;
                b := a - b;
                i := i + 1
              };
              return b}).

%% ?- fibonacci(7, P), run(P, [], _, X).

%% X = 13

pow2(N, {i := N;
         n := 1;
         while(not(i = 0)) do {
           i := i - 1;
           n := n * 2
         };
         return n}).

%% ?- pow2(8, P), run(P, [], _, X).

%% X = 256

gcd(X, Y,
    {x := X;
     y := Y;
     while(not(x = 0)) do {
       if {x < y} then {
         y := y - x
       }
       else {
         x := x - y
       }
     };
     return y}).

%% ?- A is 2*2*5*7*13, B is 5*13*7*7, gcd(A, B, P), run(P, [], _, X).

%% A = 1820,
%% B = 3185,
%% X = 455

isqrt(N, {q := N; p := (q + N / q) / 2;
          while(q > p) do {
            q := p;
            p := (q + N / q) / 2
          };
          return p}).

%% ?- isqrt(81, P), run(P, [], _, X).

%% X = 9

Discussion

As best I can tell this actually works. For instance, evaluating our Fibonacci function at 150 gives the correct value, spitting back 9969216677189303386214405760200 as should be the case. Notice how, therefore, we get, among many other things, SWI-Prolog's built-in arbitrary-precision integers for free -- a classic example of why embedded domain-specific languages are such a huge win.

The implementation depends heavily on Prolog's excellent support for custom symbolic operators and pattern matching abilities, even tolerating added bits of syntactic sugar like the brackets used to delimit boolean conditions in our if-then-else blocks.

Obviously you wouldn't want to use an interpreter this trivial for serious production/mission-critical stuff, but I think it definitely provides a compelling answer to how far a computer language could conceivably be extended to support new syntax. Languages that claim to support language-oriented programming, but don't support custom operators with custom fixity and precedence -- you're on notice.

aku tau kau masih tak sudi hubungan aneh kita ini berakhir dengan cepat.

aku tau kau pasti akan berusaha membuatku bertekuk lutut kembali padamu.

aku tau kau akan mengirimkan kakitanganmu untuk memukul genderang di telingaku.

tapi kau juga harus tau,

kau tak bisa mengalahkan pekat malam yang mengiyurkan untuk kucicipi.

kau tak dapat mengganti kasur kapuk empukku dengan secangkir kopi panasmu.

kau takkan sanggup menjadi raja di atas pelupuk mataku yang tak ingin lagi bermain  denganmu.

tapi aku yakin satu hal,

kau dan aku pasti akan berjumpa lagi.

namun untuk saat ini cukup sampai disini.

insomnia, aku akan menemui raja di dalam mimpi malam ini.

mereka gemar tertawa

mereka cinta tertawa

mereka sangat suka tertawa

mereka sangat gemar tertawa

mereka sangat cinta tertawa

mereka terlalu suka tertawa

mereka terlalu gemar tertawa

mereka terlalu cinta tertawa

mereka pikir saya badut ?

*shit !!!

bulan september, entah karena terkena sindrom bulan puasa atau apa, selama september ini, penyakit susah tidurku alias insomnia terlalu berkuasa dan terlalu merajalela. dia, memperdaya otakku, pikiranku. dia bersekongkol dengan malam dan membuat sebuah rencana agar aku tidak bisa tidur. dia sengaja membunyikan genderang dan gong di dalam kepalaku. dia menjepit kedua mataku hingga nyalang namun lelah. dia membuat jalur khusus agar aku bisa menikmati dia dengan suka cita. dia begitu memperhatikanku hingga tak satu malam pun ia lewatkan tanpa aku.

dia dan aku tak begitu sadar, bahwa kami sama2 menodai malam "disini" dengan mata nyalangku dan dia yang menari2 dengan genderangnya di kepalaku. dia dan aku menciptakan satu lintasan dimensi yang tak seorang pun tau bahwa dia dan aku sedang tersenyum.

hai, insomnia... apa kau puas setelah bermalam2 menemaniku menghirup udara, membuatku terenyuh kosong akibat kelelahan, menciptakan lingkaran hitam di bawah kedua kantong mataku, membuat perutku kembung gara2 aku hanya di temani air putih, membuatku menjumpai sahur tanpa terlebih dahulu memadukasih dengan bantal dan kasur kesayanganku, membuat sekujur tubuhku menggigil kedinginan di tusuk dengan kejam oleh angin malam, membuatku harus menghela nafas ketika menyadari semua orang sudah terlelap pulas dan bermimpi, membuat bulan ketakutan padaku dan menangis merintih hingga hujan jatuh menghantam semut - semut di tanah.

insomnia, mungkin hubungan kita memang tak bisa berjalan lebih jauh lagi. aku tak mau kehilangan separuh masa depanku hanya gara2 kau selalu mengancamku di malam hari. kau egois. kau tak mau mengalah. aku lelah. aku jengah. aku tak mau kau mengatur hidupku, menguasai pikiranku. kau itu bedebah.

jadi, bagaimana kalau  kita putus saja ?

Mungkin setelah ini, akan hiatus lagi… Gomen ne… *-*

dasar pengecut !

Esta es una Guia rapida y practica de como Diferenciar.

TAREA: DISPARARTE EN EL PIE

C
Te disparas en el pie, pero lugo te das cuenta que el arma es demasiado poderosa para tus simples necesidades de dispararte en el pie.

C++
Accidentalmente creas una docena de clones tuyos y les disparas a todos en el pie. Proveer asistencia médica es imposible porque no puedes distinguir entre las copias reales y las que señalan a otros diciendo “¡Ese soy yo, ahí estoy!”

Java
Después de importar java.awt.pie.derecho.* y java.awt.pistola.mano.derecha.* y después de escribir las clases y los métodos que necesitas, ya se te olvidó lo que estabas haciendo.

Ruby
Tu pie está listo para que le dispares en unos cinco minutos, pero no puedes encontrar ningún lugar para dispararle.

C#
Cópiale el método a Java y trata de convecer a todos que tú lo haces mejor.

PHP
Te disparas en el pie con un arma hecha con 300 piezas de otras armas.

ASP.NET
Encuentras una pistola, se desbarata. La vuelves a armar, se vuelve a desbaratar. Tratas de usar el .GUN Framework… se desbarata. Terminas mejor apuñalándote el pie.

SQL
SELECT @munición:=bala FROM pistola WHERE gatillo = ‘JALAR’;
INSERT INTO pierna (pie) VALUES (@ammo);

Perl
Te disparas en el pie, pero nadie entiende cómo lo hiciste. Seis meses después tú tampoco lo entiendes.

JavaScript
Has perfeccionado una rica y robusta experiencia de usuario para dispararte en el pie. Después te das cuenta de que las balas estan desabilitadas en tu pistola.

CSS
Te disparas en el pie derecho con una mano, luego cambias de mano para dispararte en el otro pie, pero entonces te das cuenta que tu arma se ha convertido en una banana.

Fortran
Te disparas en cada dedo del pie, iterativamente hasta que se te acaban los dedos de los pies, después lees tu otro pie y repites. Si se te acaban las balas continuas de todas formas porque no hay capacidad de manejo de excepciones.

Modula2
Después de darte cuenta que no puedes hacer nada en este lenguaje, te das un tiro en la cabeza.

Cobol
Usando una PISTOLA COLT 45, APUNTAR arma a PIE.IZQUIERDO, ENTONCES colocar BRAZO.MANO.DEDO en PISTOLA.GATILLO y APRIETA. ENTONCES regresa PISTOLA a FUNDA. VERIFICAR si las agujetas se tienen que volver a amarrar.

Lisp
Te disparas en la extremidad la cual sostiene el arma con la cual
te disparas en la extremidad la cual sostiene el arma con la cual
Te disparas en la extremidad la cual sostiene el arma con la cual
Te disparas en la extremidad la cual sostiene el arma con la cual
Te disparas en la extremidad la cual…

Scheme
(define (disparo arma pie)
(if (null? pie) ()
(disparo arma (car (cdr( pie)))))

Basic
Te disparas en el pie con una pistola de agua. En sistemas grandes, continúas hasta que toda la parte baja del cuerpo esté inundada.

Visual Basic

Tratas de dispararte pero cuando tratas de meter la bala en el arma no podes porque pesa 153Kg.

Pascal
El compilador no te deja dispararte en el pie.

Snobol
Si tienes éxito, te disparas en el pie izquierdo.
Si no lo tienes, te disparas en el pie derecho.

Euclídeo concurrente
Te disparas a ti mismo en el pie de otra persona.

HyperTalk
Colocas la primera bala del arma dentro del pie de tu pierna izquierda.
Solucionas el resultado.

Motif
Te la pasas los días escribiendo una descripción UIL de tu pie, la trayectoria, la bala, y hasta las partíclas de humo. Cuando finalmente vas a jalar el gatillo, está atascado.

Unix
% ls
pie.c pie.h pie.o dedo.c dedo.o
% rm *.o
rm: .o: No such file or directory
% ls
%

Paradox
No sólo tú puedes dispararle a tu pie, tus usuarios también pueden.

Prolog
Le dices a tu programa que quieres que te disparen en el pie. El programa calcula cómo hacerlo, pero la sintaxis no le permite explicarte.

Ada
Después de empaquetar correctamente tu pie, intentas cargar el arma concurrentemente, jalar el gatillo, gritar y dispararte en el pie. Pero cuando lo pruebas, te das cuenta que tu pie está en la tipografía incorrecta.

Ensamblador
Tratas de dispararte en el pie sólo para descubrir que primero tienes que reinventar el arma, la bala y tu pie. Después de que lo haces, jalas el gatillo, el arma hace bip varias veces, y se descompone.

Python
Tratas de dispararte en el pie pero sólo le atinas al espacio entre los dedos.

Smalltalk
Uno-mismo pie disparar

370 JCL

Mandas tu pie al MIS con una documentacion de 4000 paginas explicando como queres que sea disparado. Tres Años despues, tu pie vuelve frito.

Articulo Original

As a rule, I avoid technical (Computer Science) posts in this blog. Such posts normally appear in my other (professional) blog, http://prologsource.wordpress.com. This time, however, I make an exception. There is something very important, I think; which (if presented in a simple way) is not incomprehensible by people who are not professional programmers or Computer Scientists:

• It's about a New Sorting Algorithm, called "Tourist Sort", which is probably *the* (or at least one of the) Fastest in the world...

To be sure, this new algorithm was tested quite thoroughly, timed very precisely and compared repeatedly with many other extremely fast Sorting Algorithms. To verify this claim, you can download a demo of "Tourist Sort" by clicking HERE.

Here are some screen-shots from the (downloadable) Demo, compressed as a GIF:

The demo consists of an executable file ("TOURISTS.EXE") a brief "README.txt", an Error explanations data file (e.g. if your PC has insufficient RAM memory), and -finally- a log0file with extensive timing statistics from my PC (a Dual Core with 3 Gigabytes of RAM).

• DISCLAIMER: I'd be very wary of claiming with 100% certainty, that "Tourist Sort" actually IS THE "fastest in the world", since new sorting algorithms do emerge often, sometimes also winning prizes for being astonishingly faster than all their predecessors...

The demo is very simple, but sufficient to test "Tourist Sort" using a variety of user-generated random-data files (you can create), comparing it with two other ultra-fast Sorting programs:

• 1) An implementation of "Quick-Sort" in hand-optimized Pure Assembly Language (using code published by the "MASM project 2007", written by Steve Hutchesson).
• 2) A standard ultra-fast 'C' implementation of "Iterative Quick Sort".

So, if you download the demo and run it, you can check out the validity of my claim that "Tourist Sort" is considerably faster than the fastest of these two (typically 50% faster, sometimes twice as fast - or more).

As a matter of fact, Steve Hutchesson's optimised Assembly language version of Quick Sort (which is compared with Tourist Sort inside the downloadable demo) IS WAS probably the fastest in the world, till now... (...or at least it was among Open Source sorting projects).

There do exist faster sorting programs than this, but they tend to assume other advantages (e.g. parallel operation in special hardware). Therefore (if these considerations are correct) it is reasonable to assume that Tourist Sort is the (new) fastest algorithm, at the present time.

• Why was this algorithm named "Tourist Sort"?

Well, just as "Postman Sort" was named "Postman-" because its operation resembles the way parcels are sorted and delivered by postmen, "Tourist Sort" reminded me of the way.... tourist reservations are done by Travel Agencies, i.e.

-Instead of dealing with each tourist, individually, group-reservations are made for several tourists at once, before any tourists arrive in their destinations (and so on). Now, although this "tourist reservation analogy" is naive, simplistic, as well as theoretically flawed (to some extent), it is -probably- a useful visual hint for understanding Tourist Sort's basic idea of operation:

• In the "Tourist Sort" algorithm (an unusual descendant of Bucket Sort and Radix Sort) nothing is (actually) ever compared with anything.
  
• Forget about conventional sorting methods, forget about shuffling things around using comparisons, forget -finally- "N*Long N complexities", too...

• TOURIST SORT is of LINEAR (or quasi-linear) Time-complexity, i.e. "order of N times a constant".
  

• Tourist Sort is -however- a very unusual Algorithm (most probably a new family of Sorting Algorithms) with strong resemblances as well as strong differences with other known (rare) linear-time- sorting Algorithms. (They DO exist - you know, mine is NOT the only one...) Tourist Sort" (like "Postman Sort") belongs to a rare category of sorting algorithms that do not involve item-comparisons (e.g. lexicographic) of any kind. More details about this new algorithm's exact operation will be posted -hopefully- in my other (technical) blog, in the near future.

Tourist Sort’s History:

Tourist Sort arose out of creative experimentations with sorting methods, about a decade ago. At that time, it was implemented (rather inefficiently) in Visual Prolog. It took some years, before a working, optimised Assembly Language implementation became possible. The basic algorithm was then briefly presented in a Computer Science conference (the ALC Visual Prolog Conference) in Portugal, in April 2006. At that time, exhaustive comparative tests (performed in the offices of a company who can verify this claim) proved that “Tourist Sort” is indeed (slightly) faster than “Postman Sort”, which was a well known winner in Sorting Algorithm Speed Competitions: “Postman Sort” was proved to be the fastest Sorting Algorithm in the world, at that time. However, a year later, it was officially announced that someone else’s Sorting Algorithm had won the Annual Sorting Contest (a Chinese algorithm, in fact). (Stay tuned for more news about all this, and about Tourist Sort’s place -or purpose- in the “Sorting Contest” scene…)

• A hilarious Greek blog-story: A few months ago, a Greek blogger called "McManus" took the... piss out of my CV(!) laughing his head off, after reading there certain... wild claims I made about "Tourist Sort" (and other... strange innovations); he simply thought I was fooling around, more or less.
  
• Well, I decided it's about time that such (private) work is presented in a much better way, as well as made more open to public scrutiny and verification. Besides:

• ALL information (including new algorithms) wants to be free! :)

Some last-minute Technical tips and notes:

• When running the Tourist Sort Demo, Please avoid data-files larger in size than your RAM memory (and other factors) allow. (I haven't yet implemented the best possible use of available RAM). In my machine, files up to about 100 Megabytes are sorted OK (with occasional crashes if there are too many other windows and programs open at the same time).
• This new version of (2008-) Tourist Sort is about 25-35 % faster than the one of 2006. although I haven't worked much on improving it, because of other projects, obligations, and... blogomania. :) This (small but significant) speed increase arose from rewriting in optimised Assembly Language large parts of the algorithm that had been previously left in their original form, implemented in Visual Prolog (e.g. repeated applications of "Retract"/"Assert" commands).

Tourist Sort and Postman Sort:

Although Robert Ramey's "Postman Sort" is a work of art (as well as THE fastest Sorting Algorithm in the World, till a few years ago), and although Tourist Sort has some strong similarities with it, Tourist Sort is NOT Postman Sort; it is also very different, a new Sorting Algorithm, which... I promise to elucidate in the near future, in every detail (after a period of years during which -hoping to make money from it just like Ramey- I kept the algorithm a secret).

• Unfortunately, I was unable to run my copy of "Postman Sort", to check out if Tourist Sort is still faster than Postman Sort, since my shareware licence expired. However, this will certainly be done in the near future. (Two years ago, the test was performed several times, proving that Tourist Sort was faster - in a friendly company's offices).
• However, my strong feeling is that BIG improvements in Tourist Sort's execution speed haven't even begun; I expect a further 5-10 times speed-increase, just by rewriting crucial parts of Tourist Sort in SSE-2 SIMD (Single-Instruction-Multiple-Data) Extended Assembly Language instructions, valid in Pentium 4 and Multi-Core CPUs.
• Amusingly enough, the 2006-version of Tourist Sort would probably fail to beat Postman Sort today, if Postman Sort had been (in the meantime) re-written in optimised Assembly language.
• Tourist Sort was faster than Postman, back in 2006, mainly because Postman- was implemented in 'C' rather than Assembly. (I have no idea how P.S. is implemented today, but my strong guess is Assembly).
• Tourist Sort is also IDEAL for Parallel Computing; So ideal in fact, that most of the time the algorithm repeatedly (and wastefully) does in one CPU lots of tasks that are ALREADY completely independent, distributed, as pieces of cake for several, parallel CPUs!
  

Related articles by Zemanta

• Has there been any innovation in Computer Science in the last decade?
• Back to Basics: Algorithms
• US computer science drought may be bottoming out
• Bacteria colonies used to model parallel processing system
• reading/writing/sorting Prolog variables, using the original variable-names (LPA Prolog code)
• Assembly Language for Visual Prolog Meta-programming

In any case, some readers have given me quite a few resources on Prolog integrations and they've been very helpful.  Thanks everyone!

:- op(500, fx, many).
:- op(500, fx, one).
:- op(600, fx, always).
:- op(600, xf, notnull).

:- op(500, fx, at_least).
:- op(500, fx, at_most).
:- op(500, fx, exactly).

model(post,
      title(length(at_least 10)) notnull,
      many tags(name),
      always one author(name notnull, email(email_addr), url(url)),
      body(memo),
      many comments,
      chronological).

model(comment,
      name,
      email(validate_regex("\w+ blah blah line noise here")),
      ip(ip_addr),
      body(memo),
      nonenull,
      chronological).

get_predicate_instances(Predicate, Instance) :-
        between(1, 255, Arity),
        functor(Instance, Predicate, Arity),
        predicate_property(Instance, interpreted),
        call(Instance).

get_predicate_instances allows us to query all instances of a predicate without needing to restrict their arity by enclosing things in lists. Query facilities like it would allow you to place fewer restrictions on what kind of input a user can provide when describing an application.

The point is, if you create a domain-specific metadata language, Prolog can slice and dice it quite nicely. No XML required, no search/traversal boilerplate.

Designing a system like this in a vacuum would be a mistake; I was just toying with postfix and prefix operators more than anything else. What we need is an application we can use every day to drive the development of a system like this.

No system, small or large, should be ignored.  If it was in production once before, then it's something I'm interested in seeing.  Especially if it was used prior to 1995 and was built to process a lot of data at once.

Anything written in the following languages will do:

• COBOL
• Cognos Powerhouse
• MUMPS
• Q-BASIC/MS-BASIC
• C/C++
• SmallTalk
• Coldfusion
• LISP - which includes Autocad scripts
• Prolog
• Turbo Pascal
• Assembler
• VAX VMS command scripts
• MODULA/MODULA-2
• Ada
• PERL
• Applications written for CORBA (any language).

I've resumed work on my Honours Project.  To recap, my honours project is to create a useable Prolog Development Tool set (PGDT).  PGDT is to be used by third year students at Carleton University for the Programming Paradigms class.  Currently, any Prolog dev environment is pretty crappy however I plan on changing that.

So far, I've managed to create a working Parser and Abstract Snytax Tree that can be used by many of the other features in a typical development environment.  The AST can be used to enforce meaningful syntax highlighting (rather than just line-based rules or having expensive regex searches run everytime you save), in addition to providing and other such features.  The Editor is coming along nicely.  By modeling it similar to what the eclipse guys have done for JDT, the editor will be fairly spiffy and have a ton of features.

Unfortunately school has actually started now and as a result I'm starting to realize how bad of an idea it was to take 4 classes, my honours project and 20 hours of part-time work a week.  IN ANY CASE, there is code to write.  There will always be code to write.

OH and also, I've accepted the job that was offered from IBM.

În fapt, legăturile spirituale (şi cele care au marcat un anumit climat de creaţie) dintre Horia Bernea şi Vasile Doru Ulian s-au dezvoltat anterior grupării Prolog şi au ţinut de prietenia dintre aceştia, prietenie jalonată de momentele petrecute împreună la Văratec, Mănăstirea Neamţ, Sihăstria ...     

Un pasaj din Val Gheorghiu (Vieţile după Vasari) evocă atitudinile creatoare, în relaţie cu natura, din taberele de creaţie şi o anume atmosferă a acestora, cultivînd – poate redescoperind – avantajul naturii ca întemeiere a demersului plastic. „De cîte ori văd un pictor ieşind, cum se zice, la peisaj, mă gîndesc la mai multe. [...] Cînd îi văd, mă gîndesc cu nostalgie la epoca romantică a picturii, cînd natura pururi generoasă îi oferea artistului, dintr-o dată, două avantaje: îl învăţa să vadă şi îl desfăta la sînul ei. [...] În fond, nici o lege scrisă sau nescrisă a picturii moderne nu ameninţă că dacă vei lua lecţii de la natură, vei da greş. [...] Mă mai gîndesc la faptul că sînt pictori şi pictori. Pe unii natura îi învaţă, pe alţii doar îi desfată la sînul ei. Primii cred încă în marea lecţie a naturii şi rămîn cu ceva din această lecţie. Ei înţeleg, probabil, ce spunea Aristotel: Natura are intenţii bune dar nu le poate pune în practică. Ei îşi înţeleg, adică, rostul: de a pune în practică intenţiile naturii. Ceilalţi pictori ies la peisaj, neştiind de ce ies: aşa au făcut meşterii de altădată, aşa fac şi ei. Oricum, se simt minunat în sînul naturii. Şi mă mai gîndesc la Oscar Wilde, marele mincinos, care a pus în drepturile sale pe artistul instruit, matur, capabil să discearnă lecţia naturii. Vorbind de perioada în care impresionismul descoperea ceaţa londoneză, făcînd din ea un faimos motiv de artă, el încheia: Acolo de unde cel cult ia un efect, incultul ia o gripă.”

Printre protagoniştii taberelor, artistul instruit, matur, capabil să discearnă lecţia naturii  – reiau spusele concise ale lui Val Gheorghiu – este şi Vasile Doru Ulian.

Am extras fişa biografică a pictorului Ulian (într-o lipsire de stil, în cel „telegrafic” oarecum, în care de altfel îmi repugnă să scriu). Ţin doar la faptul că, pictorul fiind mai mereu nemulţumit de unele inexactităţi apărute prin diversele cataloage de expoziţii, am pus asupra datelor de mai la vale accentele pe care însuşi artistul mi le-a sugerat. Insista nu pe locul naşterii sale, ci pe cel care i-a definit universul copilării. Pictorul Ulian s-a născut în Frătăuţii Noi, judeţul Suceava (30 ianuarie 1944), dar a copilărit la Vaşcău, în judeţul Bihor, unde a căpătat acel special fel de a fi, ardelenesc. A învăţat la şcoala primară din Vaşcău şi apoi la Liceul din Ştei. A absolvit, în anul 1965, Institutul Pedagogic Bucureşti,  Facultatea de Arte Plastice, unde a fost coleg de an cu pictorul Horia Bernea.

Iernile şi le petrece la Piatra Neamţ însă de cum primăvara dă în pârgă, dacă nu e prin tabare, umblă pe dealurile de la Vaşcău, însoţit de o lădiţă ca de felcer vechi, în care ţine tuburile cu „magmă” policromă pe care o întinde pe pânze. Din 1968 a organizat 17 expoziţii personale (dacă are vreo importanţă răbojul) şi a participat cu lucrări la numeroase expoziţii (judeţene, bienale, saloane naţionale, expoziţii de grup) în ţară şi în străinătate. Este membru al Uniunii Artiştilor Plastici (U.A.P.) din România. În 1991 i s-a conferit Premiul U.A.P. la Saloanele Moldovei – Bacău, iar în 2005 premiul Lascăr Vorel la bienala de artă plastică de la Neamţ.

Participă – cu o contagioasă bucurie colocvială – la numeroase tabere de pictură. Lucrările artistului se regăsesc în colecţiile de excepţie ale marilor centre culturale naţionale – George Enescu, Tescani, Mihai Eminescu, Ipoteşti – în colecţii de stat şi particulare din ţară şi din străinătate. 

Vrei mai multe detalii?

Lasa-mi o intrebare si iti voi raspunde curand...

Sani dan Arli sudah dua tahun menikah. Tepatnya bulan depan. Sani adalah istri yang baik, dan Arli adalah suami yang bertanggung jawab. Sani mencintai Arli lebih dari yang Arli tahu, begitu pula Arli mencintai Sani lebih dari yang Sani tahu.

Sekitar lima menit setelah selesai memeriksa hasil ulangan murid-muridnya, Sani melangkahkan kaki keluar dari kamar menuju dapur dan memanaskan air, bersiap-siap untuk membuatkan minuman entah kopi entah teh hangat untuk suaminya nanti. Sambil menunggu air mendidih, Sani berjalan bolak balik dari ruang tamu ke dapur sambil berpikir. Sambil bersyukur, bahwa di usianya yang ke-26, dia sudah menikah dengan pria baik-baik yang bertanggung jawab, dan memiliki rumah tempat berlindung meskipun masih kontrak dan tidak begitu luas. Tapi paling tidak Sani terus berusaha bersyukur. Di saat itu lah Sani mendengar suara mobil dari kejauhan. Dia lihat jalanan dari jendela, dan dia melihat mobil suaminya. Segera setelah menemukan kunci gembok depan, Sani berlari keluar dan membukakan pagar untuk suaminya.

"Kok belum tidur sayang, aku kan bawa kunci malam ini?", tanya Arli begitu turun dari mobil setelah memarkir mobilnya.

Arli adalah sosok pria Indonesia keturunan Jawa yang manis, dengan tinggi badan yang cukup tinggi untuk orang Jawa, termasuk proporsional untuk orang Indonesia. Matanya cerah berbentuk daun, dengan tulang hidung yang cocok dengan bentuk tulang mukanya yang cukup keras untuk kategori orang jawa.

"Nunggu kamu, takutnya kamu belum makan", jawab Sani sambil memasang gembok kembali pada pagar rumahnya.

"Udah kok tadi, kenyang banget malahan!"

"Ya udah, aku bikinin teh aja, biar enak tidurnya"

Setelah mengganti baju kerjanya dan cuci muka, Arli duduk di sofa ruang tamunya bersama dengan istrinya yang sudah terlebih dahulu duduk di situ dan mulai meminum teh. Pada saat-saat seperti inilah Arli merasa sangat beruntung telah menikahi Sani. Wanita keturunan Jawa-Bali yang manis, dengan raut muka yang sangat khas Indonesia, kulit yang benar-benar kuning langsat, rambut hitam legam yang berombak halus dan badan khas penari Bali.

Pukul setengah tiga pagi, baru mereka berdua memutuskan untuk tidur dan beranjak dari sofa setelah membicarakan kegiatan mereka seharian kemarin. Arli lebih dulu berbaring di atas tempat tidur berukuran king size itu, sementara Sani masih di kamar mandi menyelesaikan ritual malamnya dengan segala macam kosmetiknya. Sebelum terlelap, Arli sejenak berpikir bahwa tidak seharusnya dia menyesali pernikahannya dengan Sani. Sementara di kamar mandi, Sani menyeka air matanya dan berkata pada dirinya sendiri bahwa tidak seharusnya dia menyesali pernikahannya dengan Arli.

"Aku sayang kamu", ucap Arli ketika Sani merebahkan diri di sebelahnya.

"Aku sayang kamu", balas Sani sambil meletakkan kepalanya di atas dada suaminya.

Pagi itu, Sani tidur dipelukkan suaminya. Baik Sani dan Arli tidur dengan rasa syukur telah memiliki satu sama lain. Mereka tidur dalam hangatnya cinta.

Tapi dalam mimpi mereka, mereka melihat penyesalan atas pernikahan mereka. Sampai dua tahun usia pernikahan mereka, penyesalan itu tetap ada dalam mimpi, tidak satupun dari mereka berani membawanya pada kenyataan. Mereka lebih suka menjalani cinta yang indah dalam kehidupan yang nyata, dan membiarkan penyesalan yang pahit dalam kehidupan mimpi.