bgj128

UK News and Tournaments ~ Tony Atkins . . . . . . . . . . . . . . . . . . . 2Ratings and Strength ~ Doing a good job ~ Geoff Kaniuk . . . . . . . . . . . 5Go Tutor ~ Edited by Charles Matthews . . . . . . . . . . . . . . . . . . . . . 6Get Strong at Scoring ~ Tony Atkins . . . . . . . . . . . . . . . . . . . . . . 8Croatian Go Song ~ Francis Roads . . . . . . . . . . . . . . . . . . . . . . . 9British championship 2002 ~ Game 1 ~ Matthew Macfadyen. . . . . . . . . 10The Way to Go ~ Andrew Grant . . . . . . . . . . . . . . . . . . . . . . . . 12Council House ~ Simon Goss . . . . . . . . . . . . . . . . . . . . . . . . . . 13European Pair Go 2002 ~ Natasha Regan . . . . . . . . . . . . . . . . . . . 14Double Extensions ~ Part II ~ Charles Matthews. . . . . . . . . . . . . . . . 17Nakade & Ishi-no-Shita ~ Part Nine ~ Richard Hunter . . . . . . . . . . . . 22Review ~ Many Faces of Go version 11 ~ Nick Wedd . . . . . . . . . . . . . 28What The Books Don’t Tell You ~ Part VIII ~ Simon Goss . . . . . . . . . . 30Sixth Isle of Man Go Congress ~ Francis Roads. . . . . . . . . . . . . . . . 35Review ~ SmartGo:Board and SmartGo:Player ~ Nick Wedd . . . . . . . . . 36Ten Years Ago ~ Tony Atkins . . . . . . . . . . . . . . . . . . . . . . . . . . 37Strange things happen in Croatia ~ Francis Roads . . . . . . . . . . . . . . 38Go Flow ~ Roger Daniel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39European Pair Go ~ UK vs Romania ~ Matt Cocke . . . . . . . . . . . . . . 42World Go News ~ Tony Atkins . . . . . . . . . . . . . . . . . . . . . . . . . 44Small Ads and Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47BGA Officials ~ Postal, e-mail and Web Addresses . . . . . . . . . . . . . . 48UK Club Lists . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49Glossary of Go Terms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

BRITISH GO JOURNAL NO 128 AUTUMN 2002 ~ CONTENTS

Sept 2002 Journal 23/10/02 10:12 pm Page 1

Golden JubileeThe Birmingham (Jubilee) Tournamentcelebrated the Queen’s Golden Jubilee bystarting the extra long bank holiday weekendoff in proper fashion. 31 players assembledalongside the 4NCL chess event at the GrandHotel in the centre of the city on 1st June.The experience of playing in a chandelier litroom will not be repeated as the hotel is dueto close before next year. The winner wasMatthew Macfadyen (6 dan Leamington)and second was Piers Shepperson (5 danCLGC). Players with 3 wins were MikeCumpstey (2 kyu Manchester) and GarethJones (11 kyu Leicester ). The Continuous9x9 winner was Peter Fisher (4 kyu Leicester) with 10 wins.

Silver Jubilee38 players took part in the LeicesterTournament on 15th June, the 25th suchevent and the tenth at the hall of the Churchof the Martyrs. The numbers may have beenlower this year because the English footballteam were endeavouring to win a World Cupgame during the middle of the day. In theGo, a top group jigo between Tim Hunt (2 dan Open University) and Simon Shiu (4 dan Bristol) meant a tie at the top. Timwas judged the winner ahead of previouschampion Simon. Phil Ward-Ackland (6 kyuWest Wales) was unusually the only playerto win all 3 games. Tony Atkins (2 danBracknell) won the most games in the13x13, despite not being registered for thetournament, and Shawn Hearn (7 kyu BerksYouth) played the most 13x13 games.

48 without loss44 players attended the tenth Welsh Open inBarmouth on 29 and 30 June. The usualmixed bag of summer weather greeted theplayers, as did new host for the tourna-ment, Phil Ward-Ackland, and of course by

Baron Allday in his cafe. Phil continuedthe well-run tradition and deserves aspecial award as smartest organiser,wearing jacket and tie for the prize giving.The winner for the tenth time was MatthewMacfadyen (6 dan Leamington). Thisincreases his number of unbeaten games inthis event to 48 (the first year the tourna-ment was only three rounds), which mustbe some sort of record. Also on 5 wins wasShawn Hearn (7 kyu Bracknell). On 4 winswere Dan Gilder (2 dan Manchester),Gerry Mills (1 dan Monmouth), DavidKing (1 kyu Swindon), Andrew Marshall(13 kyu Isle of Man) and JonathanEnglefield (20 kyu High Wycombe). TheEeyore team, comprising Simon Goss,Shawn Hearn, Emma Marchant andJonathan Englefield, won the team prize. In the 9x9 the highest winning percentagewas by Ron Bell and the Marathon Man(most games) was Roger Daniel. A specialbooby prize went to Edward Marshall toreward his travelling from the Isle of Man.

Garden Party40 players attended the first tournament atLetchworth Garden City on 13th July. Justahead of the first garden city’s century nextyear, Letchworth gained both a tournamentand a new go club as it is the new Bexfieldfamily home. The St. Hugh’s Church Hallwas pleasantly situated near the centralKennedy Rose Gardens, the museum and theshops and facilities of the centre ofLetchworth. Winner was David Ward (4 danCambridge); he beat Ruud Stoelman (2 danBradford) in the final. Organiser SimonBexfield (1 dan) was also one of those onthree wins. The others were Richard Mullens(5 kyu Stevenage), Shawn Hearn (6 kyuBracknell) and Nicola Hurden (10 kyuBracknell). Roger Daniel won the first toenter award and Jil Segerman got a prize for

UK NEWS AND TOURNAMENTS

Tony Atkins ajaxgo@yahoo.co.uk

playing all four of the games in theContinuous 13x13 competition. After theevent all the competitors were invited backto the Bexfield’s for a garden party and largepercentage of the competitors partook ofdrinks and food in the pleasant summerevening air.

Wedding Anniversary6 years to the day after the MargettsWedding Tournament at Epsom Down, thefirst Epsom Tournament was held on 10thAugust. Thanks to sponsorship of StephenStreater’s Forbidden Technologies plc therewas the fine venue of the Ebbisham Centrein the centre of Epsom and generous prizes.The modern Centre houses the town library,a cafÈ and gym and looks out on DerbySquare. The square had a statue of a jockeybut otherwise needed a water-feature tomake it complete; this was provided by thetorrential rain in the middle of theafternoon. Otherwise the day was fine andwas interesting for the players as the firstBritish Championship title match was heldin an upstairs room. Matthew Macfadyenwon this as expected, beating MatthewCocke by 12.5 points. Players were able towatch an analysis of the game after theprize giving. Close friends when theninvited back to the Margetts’ house tocomplete the anniversary celebrations. Thetournament attracted 60 players, with thetop bar at 5 dan, helped a bit by the CentralLondon Go Club being closed for refurbish-ment. Winner was Seong-June Kim (6 dan)who beat local hope Taiko Nakamura in thelast round. All players with two or morewins got a prize and all players went a wayclutching an Epsom Tournament 2002 mug.The players winning all three games were:Alan Thornton (2 dan St Albans), PaulClarke (1 dan High Wycombe), Epsom’sDavid Pemberton (4kyu), Brian Kelly (5 kyu) and Roland Halliwell (9 kyu) andReading’s Richard Brand (14 kyu) andChris Laker (19 kyu). There were two

continuous tournament winners: AlisonBexfield at 13x13 and Roland Halliwell at9x9. The Epsom Duffers won the teamprize and Gerry Mills won the first entryprize and also brought the bookshop. Sue Paterson won a prize despite notnaming Hugo the Bear correctly andnobody guessed from where the ChronosClock was imported.

Gold, Silver and BronzeThe sixth annual Mind Sports Olympiad(MSO) moved for the first time out ofLondon, but maintained the tradition of adifferent venue every year. Held over fivedays in August, Loughborough Universitywas the new home. On site parking andaccommodation were easily availablemaking the venue perhaps more convenientthan in the capital. A late start for the Goevents on Thursday 15th allowed time fortravelling. Although a weekday was conve-nient only for those without jobs, the clashwith the Isle of Man forced it so. Abeginners’ Go event featured teaching and asmall tournament; it attracted four newconverts and a child from Birmingham whohad played a little before. There was onlyone medal Go event; it attracted 18 playersfrom 6 dan to 18 kyu. Matthew Macfadyen(6 dan Leamington) took Gold for thesecond time running. Silver went to thelosing finalist Alex Selby (3 danCambridge). Piers Shepperson and FrancisRoads also won 2/3. Mark Collinson (1 kyuYork) won 3/3 to earn him Bronze. ChrisGoldsmith (11 kyu Cambridge) also won3/3 earning himself a packet of biscuits.Other than Go, there was the usualselection of games old and new. Boku,Abalone, Othello, noisy Backgammon (insilly hats) and quiet Chess. Though onlythere for one day, the Go players were wellreceived by the organisers Tony Corfe,David Levy and Lady Mary Tovey, beingdescribed as “so warm and friendly”.

Crown JewelThe jewel in the crown of the British Gocalendar must surely be the Isle of Man GoWeek. Where else can you get a whole weekof Go, seaside, countryside and pleasantcompany? The event was back after a threeyear gap and was for a third time in theresort of Port Erin. This time, however, thevenue was the disused Talon’s Nightclub ofthe Falcon’s Nest Hotel. Players mostlyarrived for the start of the week on August17th, which enabled them to eat a slice ofone of Tony Atkins’ three birthday cakes.Unfortunately there were not many visitorsfrom mainland Europe and none from furtherafield. Most welcome were father and sonfrom Portugal, Jose and Miguel Teles deMenezes, but unfortunately a family fromJapan withdrew because of illness. Also notthere was France Ellul, just leaving sevenplayers (including organisers David and LeoPhillips) who have been to all six Go Weeks.The weather was mostly sunshine and theday of the coach trip to Peel and Ramseywas no exception. Unfortunately theproposed sandcastle competition to be judgedby 21 month old Charlotte Bexfield wascancelled, but the music evening went aheadas did the American-style last evening PrizeGiving Banquet with songs after.36 players took part in the mainOpen event, which is played with90 minute time limits over fivemornings. Winner on 5/5 wasPiers Shepperson (4 dan CLGC).Second was David Ward (4 danCambridge) and third TonyAtkins (1 dan Bracknell) both on4/5. Francis Roads (4 danWanstead) was fourth andNatasha Regan (1 dan Epsom)was fifth. Also on 4/5 were Jose(6 kyu) and Miguel (14 kyu)Teles de Menezes. On 3/4 wasGary Beman (9 kyu Leamington),who unfortunately could not makethe first day. The 29-player three-

round Afternoon Tournament was a winagain for Francis Roads who beat PiersShepperson in the last round. SimonBexfield (1 dan Letchworth) and AlistairBrooks (20 kyu Swindon) both also won3/3. As usual Francis Roads also won the16-player 13x13. The next three places weretaken by Simon Goss (2 dan Bracknell),Celia Marshall (12 kyu Man) and Jose Telesde Menezes. David King (1 kyu) and EricHall (2 kyu), both from Swindon, won the12-pair Rengo. Top prize winner was youngWilliam Brooks (8 kyu Cambridge). He wonthe 8-player day-off Die-Hards’ Tournament(beating Peter Fisher (4 kyu Leicester) byjust five points). He won the Lightning,beating even younger Alex Beman (25 kyuLeamington) in the final between the toptwo players in this continuous event. Healso won the six-round 23-player Handicapevent unbeaten. David Ward and Ian Marsh(1 dan Bracknell) won 5/6 in the Handicap.Two more prizes were shared by William ashe was in the winning quiz team “The CleanShaven” (with Sue Paterson, Andrew Grant,Tony Atkins and Matthew Selby) and in thewinning “Still Cleanshaven” team in theTeam Lightning (with Tony Atkins andColin Adams).

Simon Bexfield trying to scrounge some sand fromhis daughter during the Isle of Man Go Week.

Photo:AJAx

In the Summer Journal (BGJ 127) FrancisRoads offers us (by his own admission ),some thoughts of a non-mathematical nature.These seem to have led him astray as he hasreached the rather strange conclusion that weshould try to persuade Europe to adopt ourstandard of gradings.On the basis of population alone, one wouldbe justified in raising an eyebrow or two athis suggestion. The distribution of activeplayers reported in last month’s rating list is:Russia 277; France 548; Romania 203;Netherlands 262; Germany 815; UK 332.The ordering in this list is by averageStrength over the whole population, and Ihave only mentioned countries with morethan 200 players. I do not think too many people would besurprised by this result. What is surprising inmy view is that our average strength is 0.32stones below par, and this is fairly close toGermany at 0.26 stones below par. Bycomparison Russia is 0.54 stones above par.Our Strengths are below par, but not as farbelow as people might expect. Given this evidence, one could not seriouslyexpect the whole of Europe to adapt to ourgradings when we are but 8.2% of the whole

European population and clearly below par.Interestingly, Netherlands is the countrywhich is bang on par so if you wanted astandard that would be it!Francis also does us a dis-service bysuggesting that “…kyu players who careabout this matter should enter at theirEuropean grades and shame the remainingkyu players into adjusting their own grades”.We have worked very hard to produce a fairsystem adapted to the conditions in the UK,and if he looks at the web site more carefullyhe will find a list which gives people’s UKstrengths as a guide for entering tournamentstogether with a comprehensive advicesystem in the form of a Frequently AskedQuestions page. The purpose of the list is not to shameanyone into doing anything. It is therestraightforwardly as a guide to kyu playersso that they may enter tournaments at arealistic UK Strength ( not their Europeanrating ) and consequently maximise theirperformance. It is pleasing to see that a fair amount ofadjustment has indeed taken place and thedistribution of strengths in each grade isbeginning to tighten up.

RATINGS AND STRENGTH ~ DOING A GOOD JOB

Geoff Kaniuk geoff@kaniuk.demon.co.uk

Friends of the London OpenThe Friends of the London Open is a fundset up by the Central London Go Club fordonations to cover accommodation ofEastern European, students or other needyplayers who otherwise could not afford totake part in the London Open. Prior to thetournament’s move to the International

Student House, free floor space wasavailable at the Highbury Roundhousevenue; now only more expensive roomsand dormitories are available. Anydonations should be sent to London OpenGo Congress care of Bill Streeten.

Tony Atkins

IN THE DARK?

This article begins a discussion ofthe endgame in Go. Do you regardthe endgame as difficult? Then youare in good company. For the prothe endgame is technique. What isrequired to reach dan level canundoubtedly be learned.Simple mistakes at the edge of theboard can cause large territories todisappear. Or even lose groups theireye space.In Diagram 1 Black’s side extendsfrom one corner to the other. Comethe endgame White will play 1 and3 of Diagram 2. Now Black mustuse 4 to connect. Every beginnerlearns by bitter experience to covercutting points on the second line. In this case one can imagine aplayer who has moved on a littlefrom that stage making the mistakeof assuming that Black’s stones onthe left mean White’s cut doesn’twork here. It does, and you can seehow in Diagram 3, in which Blackplays 4 elsewhere, culminating in anet. A second tactical idea which isuseful to know and works too,though not quite so spectacularly, isto make White 13 in this sequencethe atari on the first line rather thanthe net, chase Black a bit andconnect out to the left hand corner.The reader can examine this on aboard for extra instruction.The player who can accept that aminor disaster of this kind hashappened, and then put a bandageon it effectively, can save many,many games. Diagram 4 shows atypical patch. Black’s loss is keptdown to 10 points. The key play isBlack 8 there. It is not too close to

GO TUTOR: THE ENDGAME ~ THE EDGE OF THE BOARDEdited by Charles Matthews charles@sabaki.demon.co.uk

78910 11

White, and concedes somefurther profit. But Whitedoes not simply march righton in to the side, asunhappily is often seen.While Black 2 in Diagram 2is correct there as an answerto the hane move White 1,another class of mistakes of

hourly occurrence in anormal Go club comes up inthe cases of Diagram 5,where the same responseleads to a serious ko fight.One should learn that theanswers of Diagram 6purchase safety for a meretwo points.

Finally a cutting point onthe second line frequentlygives rise to disasters of the‘shortage of liberties’ sort.The example of Diagram 7is typical; Black 2 theremust be at 5. As it is theBlack corner has died.

❏ 5b1

❏ 6b

B A❏ 5a

1234 5

❏ 6a 1 23

The most common method for scoring areasunder Japanese rules is to make territoriesinto multiples of five or, preferably, ten. Anydan players reading, who do not do soalready, should practice the technique ofmaking all their territories multiples of fiveand ten during play, but as this is reallyadvanced technique it will not be covereduntil one of the much later chapters.The first chance we have to get the shapesinto fives or tens after the play has stoppedis with the filling of prisoners. In ChapterOne it was suggested that the small one andtwo point areas be filled in with theprisoners first. This is of course true, buteliminating all those with four or threepoints may also be possible given sufficientprisoners. Remember the proverb aboutstrong players always fighting kos? Well the

real reason for that is so that they have lotsof prisoners to make the counting easier.Anyway, as well as the small areas, keep aneye out for areas that are, say, 11 or 12points and slip a stone or two in them as youslap the prisoners back on to the board.Finally use the prisoners to make strangelyshaped areas more rectangular.Next comes the rearranging. Do notrearrange before filling, as this is inefficient(as is filling in a point then moving the stonelater). To start with do not move stones thatare part of the border to the other colour asyou may forget who owns the area. After abit of practice you can start to do this, butthe aim is to keep the area looking like theright colour. It is allowed to pick a stone outof the border and swap it with a stone in aneighbouring opponent’s group, but make

GET STRONG AT SCORING ~ CHAPTER 2: FIVE AND TEN SHAPES

sure the removed opposing stone is placedback into a border without changing the areasize and also avoid making the moved stonelooking like an uncaptured prisoner. If thearea is going to have a stone or two placedin the middle to reduce a rectangular shapedown to a five or ten shape, then make surethe odd stones are of the colour of the area.If you find a big area that can be madebigger, usually to a five or ten shape,without disturbing the walls, then you cantake stones out of this area and use them tofill up remaining small or awkward shapeselsewhere on the board. However do notremove too many at once, otherwise theremay be risk of a drop disturbing the position.Now what actually are the best five and tenshapes? With exactly five points the usefulshapes are ‘1 x 5’ and ‘bulky 5’ (the squarewith one ear). With exactly ten the simplestis ‘2 x 5’. The next best one you often see,and in some ways is easier to count fast asten, is the ‘3 x 4 less 2’ shape. ‘Bulky 3 withan ear’ and other exact ten shapes areusually best avoided and can usually easilyrearranged into one of the two preferredshapes. For fifteen points ‘3 x 5’ is the onlyshape and for twenty-five then ‘5 x 5’.Another fifteen that is fast to make is the ‘4 x 4 less 1’, especially if it is the corner.Twenty points can be ‘4 x 5’ or ‘2 x 10’. The ten-long shapes usually occur along anedge. The handicap dot in the middle of theside is a quick way of judging how far fromthe corner a ten-long group should stop. It isallowed to pick up the 10–4 stone whilstrearranging to verify where it is. Anothertwenty shape sometimes seen is ‘3 x 7 less 1’.For a two-high area of thirty points,remember this must stretch from one cornerup to the line before the opposite cornerhandicap dot, four lines from the other edge.For forty points along one side, then that is‘2 rows + 2’ (the extra two being round thecorner). Likewise sixty is ‘3 rows + 3’. If aten-long shape is on the middle of the side,then it is better to rearrange it into two ‘2 x 5’

areas with a couple of stones in between.Like wise a long edge group with fifteenpoints would be best as a ‘2 x 5’ and a‘bulky 5’, again separated by two stones. Often you cannot easily make all the areasfives and tens, and often of course the totalarea is not a multiple of 5. In the first case,an emergency measure is to make pairs ofshapes that add up to ten, for e x ample a ‘2x 2’ and a ‘2 x 3’, however this is only onestone movement away from two ‘bulky 5’shapes. A single stone movement, betweenareas to make some fives, is just as valid atechnique for rearranging as making that bigarea bigger as described above. Now for some shapes to be avoided, unlessinaccurate scoring is required: ‘2 x 4’, ‘2 x 6’, ‘2 x 7 less 1’, ‘2 x 9’, ‘3 x 6 less 1’,‘3 x 8 less 1’. As you are rearranging youropponents area most of these are not good asthey can be miscounted to make their areabigger not smaller, so it is not to youradvantage to try. Likewise the technique ofthree stones on two points is an increase nota decrease, so should also be avoided!

Pair Go Promotion Partners Pair Go Promotion Partners (PGPP) isa scheme run by the Japan Pair GoAssociation to get players to commit topopularising Pair Go. A red ‘passport’is issued to PGPP members listing theprinciples of Pair Go: enthusing aboutspreading Pair Go, valuing Pair Goetiquette, appreciating the pleasures ofGo. All those who play in the WorldAmateur Pair Go Championships areencouraged to join. The main supporterof Pair Go in Japan is NKB, thecompany that also owns Pandanet(IGS).

Tony Atkins

IN THE DARK?

2 Wait two hours for registrationTime to go and drink some beer;Better go and get some kunaThey don’t want our euros here.Strange things happen in Croatia,Slightly slower than last year.

3 Try to learn to speak Croatian,Seems a lot to learn to me;Seven cases and two genders,And three types of letter C.Strange things happen in Croatia,Ne govorim Hrvatski.*

4 Every second shop’s a caféAs you walk a Zagreb street.Why aren’t all those people working?Have they nowhere else to meet?Strange things happen in Croatia,Why won’t they sell things to eat?

5 Take a Go set to the café,Nice to have one quite so near.They don’t seem to want our custom,“You can’t play that board game here.”Strange things happen in Croatia,Go elsewhere to buy our beer.

6 Play the 9 x 9 one evening,Knockout is the way they play.Fourteen minutes, then you’ve finishedIf it’s not your lucky day.Strange things happen in Croatia,Why not play the usual way?

7 Off they take us to Plitvice,Nature’s beauty’s there to see.Waiting ages for the bus there,Don’t know why such things should be.Strange things happen in Croatia,We get lunch at half past three.

8 Had a great time here in Zagreb,Not too many things went wrong.Soon be time to say goodbye now,Soon be time to end my song.Strange things happen in Croatia,We’ll be back before too long.

* I don’t speak Croatian.

Black: Matthew Cocke 5DWhite: Matthew Macfadyen 6DPlayed during the Epsom Go tourna-ment on 29 June 2002.

White 6 is played to ask Black whichside he is going to push. After black 9the continuation in the corner is notso urgent. Black 17 misses the point a little.White is able to strengthen theextension on the right side whilemaking a bit of territory. Black getssomething on the lower side with 21,but this is neither a severe attack onWhite nor a secure territory.Black 23 addresses the immediateproblem of the expanding whiteposition in the upper left. the successof this manoeuvre will depend onthree things. Can Black settle hisgroup quickly? Will there still be achance to invade the top left corner?Will there still be chances to attackthe white group in the lower left?Up to 44, Black does very well atsettling his stones, and does it insente, but the other two objectiveshave failed. Black 35 at 36 looksbetter.But Black's position would not be toobad if he built up the lower sidedirectly instead of the rather perverse45, which seems to be trying toattack a strong group.White is able to patch up the connec-tion of the left side stones with 46and then get on with invading at 48.this works very well until white 72which is an outrageous overplay.White should just connect at A.Black should probably capture two

BRITISH CHAMPIONSHIP 2002 ~ GAME 1Comments by Matthew Macfadyen matthew@jklmn.demon.co.uk

171819

25262728

2930 31

353637

454647

48 4950

❏ Figure 1 (1 – 50)

535455 5657

5859 60

64656667

7374 75

808182

85 868788

899091

96 979899

❏ Figure 2 (51 – 100) 71@ 62

stones and then White can take senteon the right side for a comfortableposition.Black 73 tries to aim at both sides atonce. Having played 72, White isobliged to look after the left sidestones, and the next question is howmuch of an attack Black has at thebottom.Black 95 is big enough to keep thegame close. Next, there will be somedifficult problems for both sides inthe centre.Up to 111 both sides get cut off inthe centre. It is important in suchpositions to do some counting, so asto see which of the possible largeexchanges of groups would bedecisive. Such a possibility soonarises: Black 127 aims to gain inefficiency by threatening to kill thegroup at the bottom, but he has notcounted properly. White can ignorethe lower group and threaten Black inthe upper left before answering.Up to 132 there is a big exchange.The territorial gains are almost equal,but White does better because hesettles a dodgier group.Black tries to reopen his attack with137, but the attack runs out of steamand he needs to look for some extrasente moves on the right. Black 149is a nice try, but again his countinghas not been precise enough. Whiteexchanges groups again for afavourable simplification up to 154.After 156 White is winning by aboutthe komi. Black needs an inspiredendgame. Usually this means findinginteresting ways of not answeringyour opponent's moves. But Black answers at 157 and againat 163. When White gets 174 as wellthe game is over. White wins by 12.5 points.

1314 15

37 383940 41

❏ Figure 3 (101 – 150) 146 @ 143

58 5960

666768 69

8081 8283

8990 91

949596

101102

104 105

❏ Figure 4 (151 – 207) 186 @ !

It was some time in February 1976,and I was sixteen years old. I waswalking along a local shopping streetwhen I saw something in the windowof a second hand furniture shop. Now,please, I don’t want anyone to get theidea that I was the sort of kid whofound furniture interesting, but I wasbored, I wasn’t going anywhere inparticular, and I’d been glancing at allthe shops along the street. When Ilooked in this particular shop window,though, something caught my eye.This shop had set out a coffee table anda few chairs as a display in thewindow. There were some things onthe table, meant, I imagine, as props tomake the display look more homely – acouple of books, a teacup, and – thisbeing what drew my attention – a greencardboard box bearing the words ‘GoGame’, with a picture of two ancientChinese people playing this ‘Go Game’(whatever that was) on a tree stump,and a diagram showing some black andwhite circles on a grid of lines.I looked at it with some puzzlement,for I had no idea what “Go Game”meant. Obviously some kind of game,but “Go”? The only game I knewcalled “Go” was a travel gameproduced by Waddingtons (it’s called“Travel Go” now, but it was just “Go”then), but this was clearly somethingdifferent from that. And yet I had thefeeling I’d come across it beforesomewhere.I was on my way home when I remem-bered. I was interested in board gamesgenerally, and had been looking at theChess books in my local library a fewweeks earlier. Naturally all the board-

game books were grouped together,and I now recalled seeing a book thereentitled “Go: A Guide to the Game”.I’d given it no thought at the time, Ihadn’t even bothered to pick it up –you can hardly look at every book yousee in the library, and I was looking forbooks on Chess at the time anyway.The next day I went back to the libraryand took a look at this Go book. Therewas no doubt it was about the samegame I had seen in the shop, so Iborrowed it, plus a two volumebeginners’ book from the Nihon Kiinwhich was the only other material onGo they had. By the time I got halfwaythrough the first volume of this, I knewI was going to have to buy the game.I went back to the shop and wassomewhat brusquely asked by theowner what I wanted. (I suppose theydon’t get very many sixteen-year-oldboys in furniture shops.) Trying not tosound too keen (in case it affected theprice), I explained that I was interestedin the game in the window, and wouldhe consider selling it to me? “Ofcourse”, I said, “I know it’s a prop foryour window display but...”Once the owner realised I wasn’t aboutto rob the till, he was quite happy tosell me the set. He explained that itwas a new Chinese game (which mademe smile a bit for I’d already read thatit was 4000 years old), and offered tosell it to me for a fiver. Unfortunatelymy pocket money didn’t run to this, butI asked if he could hold on to it untilnext Saturday when I’d have enough tobuy it. He agreed to this – he hadn’tbeen expecting to sell it to anyone,anyway – and I went home.

THE WAY TO GO

The next day, impatient to startplaying, I made myself a temporary set– I drew a 19x19 grid on a bit ofcardboard (none of this small-boardrubbish for me, I grandly decided) andcut some black and white card intosquares to serve as stones. It wasn’tvery good – its main disadvantage as aGo set was that you had to hold yourbreath while playing or the “stones”would go flying everywhere. Still, thiswas my first Go set.

Saturday came, and I finally bought thegame in the shop. It was clearlysecond-hand, and a bit beaten up, but itwas a real folding magnetic set onwhich you could play without fear ofoxygen starvation. But best of all, therewas a leaflet from the BGA inside it.At last I had some contact informationthat would allow me to find a local Goclub. I wrote to the BGA – and neverlooked back.

Andrew Grant

New AppointmentsWe are delighted to welcome Nick Weddinto his new role as the BGA PublicityOfficer. In Council’s view, publicity is oneof our most important functions, and wehave wanted to increase the scope of thisrole. So we have added to the existing briefsome further items that encourage thePublicity Committee to initiate new projects,which we shall support with funds whereappropriate.

The full brief is now:

● to maintain our national press contacts

● to advise clubs on publicity as requested;this includes the existing function ofadvising on how to prepare pressreleases, but is not limited to that

● to generate new ideas for local-levelpublicity

● to generate new ideas for publicising boththe BGA and the game of Go

Please would eveyone with ideas about howto enhance our publicity please contact Nickto suggest them.This means that Nick moves on from hisprevious role as BGA webmaster, and we arealso delighted to welcome Allan Crossmanas the new webmaster.Over many years, Nick has built the BGAweb site into one of the best Go-related websites on the internet. Perhaps the very best ofthose run by national Go associations. Hedeserves our warmest thanks for his skillfuland diligent work.Council wishes Allan and Nick everysuccess in their new positions.

COUNCIL HOUSESimon Goss simon@gosoft.demon.co.uk

The 5th European Pair Go championshipswere held in the Palais des Festivals inCannes. This grand building overlooking thesea has been host to the Film Festival,computer exhibitions and numerous businessconferences but in the week of 27th Feb tothe 3rd March 2002 it was dedicated to amassive games extravaganza.There was chess. The first NAO masters wasthe strongest closed tournament ever to beheld in France, boasting 7 of the world’s top20 players and including ex World championAnatoly Karpov. Once spectators managedto find their way to the theatre they were infor a treat. The seats were plush andcomfortable and there was one of the bestdisplays of the games I have ever seen. Theboards were on the stage but also fedelectronically to a cinema-sized screen. Halfthe screen gave close-up shots of the playersand the other half diagrams of the positions.There was headphone commentary availableand bulletins produced after every round.Live cinematic coverage also glamorised theScrabble, with the French championship andthe world Francophonic championship beingfought out to a packed arena.

Bridge was played in a room with a superbview across the marina. There was a largemarquee for model railway and airplanedisplays. Children were amused by the discodancing competition. There was Othello,backgammon, oware, draughts, fingerbilliards, tarot, Chinese whispers (in French)and many many more. My favourite to walkby were the fantasy gamers who dressed upin capes and grew their beards very long.The Go was supported by the European andFrench Go federations and the Japan Pair GoFederation. 10 European countries took partand the countries and ratings are shownbelow. The tournament was a 5 round Swiss;45 minutes per pair sudden death using Ingclocks but Japanese counting. Komi was 5.5.Seedings were done using European ratings,which affected the draw for the first round –thereafter it was random.(Geoff Kaniuk willbe pleased to note that our average ratingwas the highest (!!) of the3 dan pairs –although I’m afraid to say we weren’t luckyenough to finish highest.)The English contingent consisted ofMatthew Cocke and me to play, withMatthew Selby to keep us company.

COUNTRY RATING AV. EURO FEMALE MALE

Norway 4 dan 2369 Xia Jie Li 3d Xia Jie Lin 5dGermany 4 dan 2358 Daniela Trinks 3d Lutz Franke 5dUK 3 dan 2297 Natasha Regan 1d Matthew Cocke 5dFrance 3 dan 2265 Myrtille Cristiani 1d Paul Drouot 5dCzechia 3 dan 2200 Lenka Dancova 3k Vladimir Danek 6dRomania 2 dan 2152 Irina Patricia 2k Mihai Petra Bisca 5dNeth’lnds 1 dan 2074 Marianne Diederen 1k Peter Zandveld 4dSlovakia 2 kyu 1890 Lucia Lassakova 2k Ivo Svec 2kBosnia 2 kyu 1883 Visnjica Barisic 5k Dragan Barisic 4dItaly 10 kyu 1125 Francesca Antonacci 5k Emanuele Cisbani 12k

EUROPEAN PAIR GO 2002Natasha Regan natashar@aol.com

Our journey can only be described as one ofincreasing excitement starting as it did withthe Stansted ‘Express’, all change at BishopsStortford. We flew Go from Stansted toNice. I have to recommend Go. Not onlywas it the cheapest airline on this occasion,but we also learned the catchy “Go go go gogo!” jingle on the inter-terminal shuttle. Tryit – you’ll be humming it for weeks.At Nice airport we first saw the ‘FestivalInternational des Jeux’!17posters and then again at the train station.

The posters were all over the place. We tooka double decker train to Cannes and had theadditional thrill of fare evasion because theticket office closed just as we reached it andrefused to sell us anything at all.We arrived on Thursday evening; the mainevent was starting on Saturday. To preparewe ate out at Chez Freddy’s and learnt hownot to eat artichoke. We climbed countlesssteps into the sky and saw the castle androunded off the evening at Jimmy’s, thelocal casino.For months Piers Shepperson has been tryingto persuade us to take up poker. Of coursewe won’t, but we did have measured successagainst the slot machines. They kept payingout. We had to use a bucket to collect all ourEuro tokens. In the end we got so bored weshovelled it all back in and ended up twentyquid down on the night.Friday was a boat trip and then manicallyfast 10 minute blitz Go. Matthew C’s firstround was White against the 2 dan profes-sional, Fan, who is currently living in Paris.Fan was around to give commentaries on thepair Go, with help from his sidekick Pierre.Pierre would suggest a move with a flashtactical refutation at which point he’d hit hishead. Fan developed the catchphrase “Tryagain, Pierre”. Anyway in the blitz Fan beatMatthew once, then again giving 2 stones.Not surprisingly he went on to win thecompetition.

Later I played the pro’s girlfriend, who hadbeen taught the moves only two days before.I did manage to win but I don’t want areplay next week.Our meal out on Friday was the traditionalFrench (?) dish of paella. Matthew S did afine managerial job of fishing out andfeeding us the biggest cloves of garlic tomaximise our winning chances the nextmorning.This did not work however. We played theRonamians in the first round and they playedwell. We had good chances to kill a groupbut it survived and we lost by around 10points. Things went from bad to worse whenboth my companions ordered quiche forlunch.The next game was against Italy. On paperwe had a clear advantage and we startedmaturely by gaining territory throughfencing in and pressurising their groupswithout killing them.Then horror. Matthew obviously felt heneeded to prove his manhood after thequiche-eating debacle and went to overplaycity on the top right-hand corner of theboard. We recovered ourselves with the lossof only one large corner and luckily ourearlier territory and time advantages provedsufficient.Round 2 saw the top two seeds clash withGermany emerging victorious over thestrong pair from Norway. Only France,Germany and Romania remained undefeatedafter 2 rounds and the home crowd wasdelighted when France beat Germany to endthe day as sole leaders.Our third game was against Holland repre-sented by Marianne and Peter, a very nicepair who work in a Go shop in Amsterdam.As usual we got into a fight but my partnermanaged to cleanly secure victory with atrick to capture the crucial cutting stones.This secured us a place on top board the nextmorning against the leaders, France

We had another restaurant expedition in theevening and this time we ate sauerkraut. Wewere amused when the organiser, Anne cameto each person in turn in a state of panic toexplain the “bad news”. She hates sauerkrautherself and had intended a typically Frenchmeal!Then the real preparation began – backgam-mon and actuarial exam questions. As youmight imagine there’s hours of fun to be hadin reading aloud questions beginning “Youare a general insurance actuary…”On Sunday morning we had White againstMyrtille and Paul from France. Start time10am. Security to the building was so tightand crowds so large (really!) that wecouldn’t get into the building until 10.40,when play eventually began. David, a 5 danfrom Paris, was recording our moves andFan the 2p watched until my excruciatinglybad hane sent him to sleep. Nevertheless, we managed to fence off quitea large central area while they built up onthe left. We invaded the left side and Iwasn’t sure where it was all leading untilMatthew masterfully came up with a plan toconnect out. I was filled with admiration.(Do you like this phrase? I have lifted itfrom Takagi’s commentary of Otake-Kobayashi Nov ‘92, where it appears twice.)We didn’t know it yet but we would be theonly team to beat France who would go onto win the tournament. If only we could beatGermany too…Our final game was tense. 45 minutessudden death is pretty fast in an individualmatch but when you have to fathom out yourpartner’s moves as well as your opponents’the time flies by. Although there were onlytwo games on the Sunday (less than threehours total playing time) I finished the dayabsolutely exhausted.We started well against the Germans in theChinese fuseki. They made two invasions onthe right side of the board which would havebeen attackable had I kept them separated.

However they connected and we had toinvade the left side in return, but our grouprapidly became dodgy.Matthew saw a crafty way of making it livebut I couldn’t see what he was up to. Ithought ‘to hell with it, let’s play some pairgo’ and tried to rescue some rather unlikelystones in the middle. Remarkably both ourplans were successful and we emerged witha great position. In fact Lutz asked Danielathe ‘Shall we resign’ question. Annoyinglyshe did exactly what I would have done andsaid ‘play your move and quickly!’Each team had only seconds left and 3 of the4 players knew exactly what was going on.UK were killing a large group of Germany’sand there was nothing they could do. Teehee. Sadly it was me who did not realize andto this day I do not know which stones wewere capturing, though I understand therewere lots of them. So they escaped and welost.Norway-France was less clear. It couldn’thave been a closer end to the 5th EuropeanPair Go Championships as nobody knewwho had won. The game ended with a halfpoint result, until one of the players discov-ered a prisoner on the floor. That old trick!The whole game had to be reconstructed: notan easy matter when there was a long halfpoint ko fought out at the end. There waslots of dust but when it had cleared theFrench had won that match and also thetournament on tie-break too.

FINAL POSITIONS

1st France 4/52nd Germany 4/53rd Romania 4/54th UK 3/55th Norway 3/5rest =

It is a few years ago now that I firstencountered my co-author Seong-june Kim. I was summoned by phoneand properly introduced by anotherKorean player, in the cybercaféwhere the Cambridge club wasstarting to put down roots. In thosedays Seong-june had rather lessconversational English, and weplayed in quite a serious atmosphere.He offered me four stones, somethingI’m glad to say never happenedagain. The game started as inDiagram 1.The plan with Black 4 is to take thebig point at 6. If Black plays 4 at A,White will have the chance to get toeither B or C next. While this isn’tnecessarily bad for Black, I wascontent with 4 on the third line,feeling that White could claim noadvantage gained. The doubleextension that is my topic appears onthe top side with 1, 3 and 5. A more recent discussion withSeong-june revealed that Diagram 2is something he considers quiteplayable for Black. The points 2 andD being miai, Black is entitled toplay at 1. Then up to 5 would bequite normal. His comment is that themiai are also ‘miai for Black’ in the(possibly) extended sense that Whitecannot really hope to play at both ofthem. That’s a judgement on White’srequirement to take care of the groupon the left side before advancing tothe right side. In any case the usualinterpretation of miai is that youdon’t mind allowing the opponentfirst choice out of a pair of points,and Black 1 says that loud and clear.

DOUBLE EXTENSIONS ~ PART 2Charles Matthews charles.r.matthews@ntlworld.com

I want to concentrate on one way ofplaying against the double extensionthat has performed well for me, notjust in this game but against pros inteaching games. The idea is to play asin Diagram 3.In this sort of position it is large forBlack to close a corner at A or B, ifyou consider territory alone. Ifhowever there is fighting going on,you should consider a play like Black1 there. In handicap Go there usuallyis some central combat, as White triesto get into the game. Therefore I wantto explain the well-hidden direction ofplay behind this side contact move.The basic sequence of Diagram 4 is apattern worth committing to memory.Naturally Black 9 depends on havinga favourable ladder. If you assess theresult:1 With a good ladder, Black gains

substantial central influence whileWhite’s capture of 3 is quite small;

2 This ladder runs into a part of theboard where White may not haveany really scary ladder-breaker,because Black’s upper right cornercan stand up for itself pretty well.

3 White’s double extensionformation no longer looks liketurning into much territory.

4 If the ladder with 9 isn’t currentlygood for Black, Black will be ableto play a ladder-making move onthe right side instead.

It is point 4 that catches the interestwhen one holds the black stones in ahandicap game. As a plan ofcampaign, allowing White to start afight on the right side while schemingto insert a ladder-related play at oneof the ‘x’ points in Diagram 5 issomething that works pretty well.

I say this having tried it out on bothFeng Yun (then 8 dan) and Magari9 dan while they were visiting theUK, the latter game being inanother cafÈ during 20 minutes aswe waited at Paddington for theBristol train. For once, it seemed tome on playing this way, Blackescapes the feeling of followingWhite’s lead round the board. Asyou may have gathered from thetone of the first paragraph, Imanaged to land a punch on Seong-june this way, too.Based on those experiences itseems that White most likely sortsout the ladder aji by playing at 1,and allows Black to do his worstafter the ladder-making play. Theprocess is therefore much like a kofight, in that Black gets two playsin a row elsewhere in the board.One must of course factor in thecost on the upper side when comingto an overall judgement. Black’schances of invading deeply therehave gone. The two marked stonesaren’t ridiculous, though.One comment about the earlier playthat should be made. In Diagram 7White’s capture with 1 is tempting,to deny Black a later good play atA. It isn’t however sound from thepoint of view of technique. Black 2threatening a ko on the edge willnormally be troublesome. The morepatient play with 1 at 2 is thehonest way. Strong players paygreat attention to closing-off plays,and White’s stone on the rightneeds all the assistnace it can get.It seems that if White dislikes theladder-based combination, somebranch must be found back at thebeginning. Here are a couple ofexamples from pro play.

The game in Diagram 8 is GoSeigen (B, then 7 dan) againstOnoda Chiyotaro 7 dan fromthe 1942 Oteai. White connectssolidly with 4, hoping to deallightly with the upper right. Iwouldn’t venture to suggest Icould tell you what Black wasthinking about with 1; but whynot read out the implied ladderfor yourself? Black cuts with13 to take advantage of goodshape and solid positions.In the continuation Whiteconnected along the top sideand Black played 6 as a round-about attack on White’sfloating stones. White 9signifies both an exchange ofcorner territory for the side,and White’s plan of nextmaking life in the centre. Blackwasn’t in that much troublewith his weak group afterWhite pulled out the cuttingstone (6 of Diagram 8), andwon by resignation.The second example showshow the side contact play mayalso work in a differentcontext. The game is a victoryby Michael Redmond (8d atthat time) with White againstCho Chikun 9 dan in the 1998Oza Tournament. Strategicallythis game hinges on White’sright-side framework, and afterWhite 1 what is at issue iscentral influence. ThereforeBlack 2 is natural enough, andWhite embarks on manoeuvresto force Black’s hand. White 7cuts off a single black stone, atleast for the present.

❏ 10

Diagram 11 shows how White took iton from there, putting in place alarge framework strategy up to 20.White 8 and 10 contribute: they dohave to be seen in the light of furtherforcing plays remaining to Whiteover on the left side that make use ofthe sacrifice stones. Black had toresign this game when a group fillingup much of the right side came to asticky end.Other responses to the side contactplay are certainly seen: there’s theother hane ‘on top’, and the othersolid extension towards the edge.They respectively start a fight andgive away influence, and on thewhole count as less complex.Something about the pattern ingeneral is in Attack and Defense,Chapter 8, without the doubleextension setting.

141516

❏ 11

BGA PUZZLE AND QUIZ BOOKThis puzzle and quiz book is dedicated toJohn Rickard, who died in May this year.John was regarded as a mathematical geniusat school and always loved solving puzzles.All the contributions have been used at Goevents and the book has been compiled byTony Atkins.The book is avaailable from BGA Booksand all proceeds will be donated to theBritish Liver Foundation.

The previous article in this series ended withtwo problems. The first was rather difficult,but very interesting. Its many variationsreviewed several of the concepts we havelooked at, not only in the previous article,but in the whole series to date. If youthought it was easy, then you failed to findthe strongest moves for White. This problemis full of ‘katte yomi’ pitfalls.

Diagram 1There are several vital-looking points: A-D.Other moves, such as pushing in above B or playing atari on the left, are all easilyrefuted.

One useful technique for solving life-and-death problems is to visualise a desirablegoal instead of considering moves one at atime. Diagram 1a adds three black and threewhite stones. This white group is dead, aswe have seen before in the nakade half of

this series. So can Black achieve this result?Which of these three moves should he startwith? Does it make any difference?Let’s start by looking at A in Diagram 1.This move only looks good if you expectWhite to answer feebly. In Diagram 1b, after

Black triangle, White 1 at 3 or 2 offers littleor no resistance. If White connects at 2,

Black 1 leads straight to his goal of Diagram1a. White 1 at 3 will lead to a ko, which isnot good enough either. However, White 1 inDiagram 1b completely refutes Black’sattack. Black has no better followup than thesnapback, but it’s an empty threat. Capturinga square four doesn’t stop White frommaking a second eye by means of ishi-no-shita. White 1 may look like bad shape whileBlack triangle looks like a snapback-threat-ening tesuji played on a vital point, butappearances are deceptive. Vital-lookingpoints are merely guidelines for whichmoves to read out. What’s important isreading out to the end and finding the bestmoves for your opponent as well as foryourself. This is not a one-move problem.Black must read a bit more deeply than this.Diagram 1c: Starting at 1 here lookspromising, but it quickly runs out of steam.Once again, Black’s threat of a snapback is

not strong enough to stop White frommaking eyes. White responds with 4 andlives unconditionally.

NAKADE AND ISHI-NO-SHITA ~ PART NINE: SEEING UNDER THE STONES

Richard Hunter hunter@gol.com

❏ 1 Where to start?

❏ 1a Possible goal

❏ 1b Ishi-no-shita

❏ 1c Fails

Diagram 1d: Black 1 is the third vital point.If White 2, then Black 3 fails. Black 5 is

answered by White 6, not by a connection atA, which would lead to Black’s goal ofDiagram 1a. After 6, the result is effectivelythe same as Diagram 1c.Following Black 1 with 3 in Diagram 1efails too. White takes the vital point of 2again and lives. Next, if Black A, White B.

After exchanging 1 for 2, Black 3 inDiagram 1f is correct. This may seem arather unlikely move, and indeed if youhaven’t seen it before, you may welloverlook it.

It looks as though White 1 in Diagam 1gwill make one eye on the left and a secondeye on the right by ishi-no-shita. But this iswhere the 1-2 exchange in Diagram 1fshines. Black pushes in at 2 and ataris at 4.

Finally, Black captures the square four with6. So why can’t White just play his ishi-no-shita and make a second eye? Because Black6 puts all the white stones on the left inatari! White doesn’t have time to cut underthe stones. This is a key technique in theishi-no-shita repertoire. We saw an exampleof it in the last part, so this is a good review.What if White connects at 1 in Diagram 1h?This time, pushing in at Black 3 will not

work. Instead, Black exchanges 2 for 3 andthen extends at 4. This catches White in ashortage of liberties. He would like to crushBlack by playing at A, but that’s self-atari. IfWhite does nothing, Black B is atari. SoWhite’s only choice is to capture the twostones with C. This allows Black torecapture at 4 and the final shape (notshown) is essentially the same as Diagram1a. Black has contrived to play on all threevital points and make a partial eye. The two-stone sacrifice is a useful technique to know.What if Black plays 1 and 3 in the reverseorder, as in Diagram 1i? Well, this wouldonly work if White blocked to the left of 1.But instead, he connects at 2. Now he can

answer Black 3 with 4. After 6, White willbe able to play at A and make a second eye.The 1-2 exchange proves to be very bad forBlack.

❏ 1d White lives

12 34A

❏ 1e White lives

❏ 1f Correct

❏ 1g No time to cut

❏ 1h Liberty shortage

❏ 1i Wrong order

If White answers 1 with 2 in Diagram 1j, hedies. Black extends to 3, forcing White 4,and then extends again to 5. Note that 5 at 7

fails: White doesn’t capture, but secures aneye on the left with 5, leaving an ishi-no-shita on the right. White 6 is the vital pointto prevent Black from making a nakadeshape, but Black unleashes his snapbackwith 7. White is short of liberties on the left,so he can’t atari above 1. White 8 is anemergency measure that sometimes producesa ko, but not here. Can you read out theresult?Diagram 1k: Black captures five stones with1 and White cuts under the stones at 2. Black3 is necessary, but after 5, the result is notko. White is dead.

Seeing Under the StonesThe theme of this part is seeing under thestones. How good are you at visualising theshape that will be left after stones arecaptured and removed? Moves and sequencesthat seem obvious in the open can be quitehard to see when there are stones in the way.

Diagram 2Black to play. Which is better A or B? BlackA is sente; it forces White to capture.Although White captures a dogleg four, thatdoesn’t make him alive because he has aweakness in his shape.

Black’s cut at 1 in Diagram 2a is atari, butthis does not kill White unconditionally.

Rather than connecting and allowing Blackto extend to 2, White plays 2 and fights a ko.Actually, Black doesn’t need to rush toconnect at A in Diagram 2, because theresult will be effectively the same if Whiteplays first. The connection at A doesn’taffect the life-and-death status.Black’s connection at B in Diagram 2 is alsosente. White captures six stones this time,but what’s important is the shape. Next,Black’s peep at 1 in Diagram 2b is no good.

We had an example of a successful peep inthe last part (Diagram 11b), but here theshape is different.Here, Black 1 in Diagram 2c is the killingmove. White has a weakness in his wall atA. For convenience, let’s imagine that thereis a white stone at A. Then Black 1 is clearlythe vital point of White’s five-point eye-space. Once the space is vacant, this is easy

34 56 7

❏ 1j Ko?

1 234 5

❏ 1k Dead

❏ 2 Which connection?

❏ 2a Ko

❏ 2b Don’t peep

to see. Seeing it in advance, when there arestones obscuring the position, is harder andtakes practice. What about the weakness atA? Well, it does nothing to help White; evenif White connects at A, Black 1 kills him.The confusion arises in Diagram 2. Blackknows the bulky five nakade shape by heart,but connecting at B means that Whitecaptures six stones—a non-nakade shape.The key here is to ignore the hane stone onthe edge (which becomes A in Diagram 2c)and regard the six-stone sacrifice as equiva-lent to a bulky five. We looked at a similaridea when we discussed ‘pushing into thehane’ in part 3 (BGJ 122). CompareDiagrams 2 and 2c and try to see the shapeunder the stones.

Diagram 3Black to play.

If Black does nothing and lets White capturehis six stones by playing the marked stone inDiagram 3a, White will get a living eye-space. Next, Black would have to play A tostop White from playing there, but White Bmakes a second eye for the group.

So in Diagram 3, Black should capture thewhite stone in atari. This may look silly,because it gives White a snapback. But whatis important is the shape of the capturedstones. By sacrificing an extra stone on thefirst line, Black creates a weakness inWhite’s wall, as shown in Diagram 3b. Now when Black plays 1, White B doesn’tmake White alive; Black can throw in at A.And if White answers 1 by connecting at A,

Black extends to B, which will lead to deathby nakake.Here are some positions for you to practiceseeing under the stones. Reading Problems1–VI: All are Black to play.

❏ 2c Nakade

❏ 3 Black to play

❏ 3a Alive

❏ 3b Dead

Reading ProblemAnswersI. Black was in atari, so hehad to capture the five whitestones with the triangledstone. Once the stones have

been removed, White cuts at1, which is atari. Black doesnot connect, but plays 2 andfights a ko. It might look asif White 1 at 2 is the vitalcentral point, but Black 1makes two eyes.II. In the original position,Black captures the singlewhite stone and White

retakes, capturing six blackstones. Next, Black 1 is thevital point of White’s eye-space. This time playingatari immediately with 1 at3 is no good, and Black 1 at2 also fails. In both cases,White lives by playing 1.

With 2, White makes a ko.White 2 at 3 dies uncondi-tionally. All this is easy tosee when the space is vacanthere, but not so easy to seein the original position.III. After Black captures thesingle white stone andWhite retakes, Black cuts at1. Next, White 2 is a good

move; playing 2 at 3 diesunconditionally. After 5,White has a ko for life.IV. If White captures thefour black stones, in theoriginal position, he gets aflawless straight four, whichis alive. But if Black

captures the white doglegfour, White cuts in the spaceleft behind (White triangle),which is atari on the fiveblack stones. But this shapehas a weakness. Blackpushes on the first line,forcing White to take thestones off (diamondexchange). Once the stoneshave been removed and the

❏ I Ko

1 23 4

❏ III Ko

❏ IV Dead

❏ II Ko

❏ III

❏ IV

❏ VI

❏ II

space is vacant, the result iseasy to see. Black 1 threatensto extend at either A or B.V. After Black captures onthe 1-1 point, White cuts at1, which is atari on the twoblack stones. Instead of

connecting at 3, Blackcounter-ataris with 2. WhenWhite captures with 3,Black throws in with 4 attriangle; this recaptures thewhite stones in a snapback,so Black gets two eyes.Note the importance of theopen liberty at the 4-5 point.VI. After Black captures thesingle white stone andWhite retakes, Black cuts at

1. Now there is no way forWhite to make a secondeye, even in ko.

Diagram 4Now let’s discuss theanswer to Problem 2 in thelast journal. Black can makean eye in gote with 1, butWhite 2 stops him getting asecond eye. Black 3 is

easily handled by theconnection at 4.Black 1 in Diagram 4a iscertainly forcing. Whitemust play 2 to steal the

other eye. Next, Black 3looks like a vital point (if 3at 4, White answers at 3); it seems to make miai of 4and 5. However, it doesn’twork. White connects at 4and sacrifices his stones.After 5, the position is thesame as the one we juststudied in VI above, withthe colours reversed.Black should start with 1 inDiagram 4b. Vital pointsoften have to be played withthe right timing. Next,White 2 is too simple. Black3 leaves A and B as miai.Compare this with Dia. 4.

White should play 2 inDiagram 4c. Next, Black 3is a good move, forcingWhite 4 (Black 3 at 4 leadsback to Diagram 4a). Now when Black plays 5,White must fight the ko at 6to open up a liberty.As usual, I’ll end with twoproblems for you to studybefore the next part. Bothare Black to play. Problem 1 reviews someideas covered in this part.(Hint: move 5 is the key.)Problem 2 introduces thetheme of the next part.

❏ 4 Fails

❏ 4a Fails

❏ 4b Black lives

❏ 4c Ko is correct

❏ Problem 1

❏ Problem 2

❏ V 4 at ! Alive

❏ VI Dead

In issue 109 of this Journal, I wrote a reviewof version 10 of David Fotland’s programThe Many Faces of Go. Since then, I havebeen recommending it as the best all-roundGo program for someone who is willing topay for the best. Version 11 has now beenreleased. Like version 10, it is a program forWindows, and comes on a CD.Many Faces of Go is primarily a Go-playingprogram – it is one of the world’s strongestsuch programs. In addition, it comes with alarge amount of other useful material, andmany Go players will value it for this, ratherthan for its playing ability.

Playing AbilityMany Faces of Go version 11 (MFoG11) cannow play on any sized board from 7x7 to19x19. It can play against a human, as Blackor as White; it can play against itself; or itcan act as a board for a game between twohumans. It can be set to use American,Chinese, Japanese or SST (Ing) rules.It plays as fast, or as slowly, as you tell it to.Unlike many computer Go programs, itadapts its speed of play to use the timeavailable to it – and unlike some, it playsbetter if given more time. It can play veryfast. I set it to play itself on a full board atits lowest strength level, and the entire gametook 10 seconds.I tried playing against it at the highest of itsten levels of strength, giving it nine stones,and beat it (I am about 1-kyu). This issomething which I still cannot do againstHandTalk. The reason for this difference is, Ithink, that Many Faces plays ‘honestly’,whereas HandTalk tries sequences which, ifplayed by a human, would be described asswindles.I also tried playing it against a recent versionof Mick Reiss’s program Go4++. The gamewas close, but the styles of the two programs

are very different. Many Faces plays ratherlike a human 10-kyu, making influence andthen not knowing quite what to do with it.Go4++ behaves as if it knows its ownlimitations: it contents itself with modestthird-line territory, and relies on the fact thatits opponent won’t know how to make themost of its outward influence.Commercially available programs such asMFoG11 are ‘frozen snapshots’ of theirprogrammers’ work, as this is undergoingcontinuous improvement (or rarely, relapse).MFoG11 was frozen shortly before therecent 21st Century Cup, which it won,defeating all its opponents, includingGo4++. It can therefore claim to be thestrongest computer Go program that you canbuy. I doubt that it is measurably strongerthan the latest version of Go4++. But itshould be stronger than the currentlyavailable version of Go Professional, whichis a version of Go4++ frozen about twoyears ago.

Other FeaturesAs well as playing Go, MFoG11 includesmany features not present in other Go-playing programs.There is an introduction to Go, starting at avery elementary level, explaining liberties,connections, capture, and so on. This isdisplayed more clearly on a computer screenthan it could be in a book. There is plenty ofadvice on play, and two sample games withvery full comments directed at beginners.There is a Joseki tutor, and two largedatabases of openings which can be used forstudying fuseki and joseki. It can analyse thestatus of groups. It includes a problem-solver, which is much stronger than the onethat was included with MFoG10.You can use it to record and replay games,and to add variations and comments to the

REVIEW – MANY FACES OF GO VERSION 11Nick Wedd nick@maproom.demon.co.uk

game record. Something which I find partic-ularly useful is that it can record games ineither Ishi or SGF format, and can read bothformats. It can therefore be used to convertbetween them.It comes with two large collections of gamerecords: 1200 commented amateur games,from the Go Teaching Ladder, and over 1600commented professional games.It also comes with a collection of over 2000Go problems. It keeps a record of yourperformance at these, so you always have anincentive to try to get them right andimprove your score. When you ask it to setyou a problem, it randomises the orientation,so you get rather better value than if theywere always presented the same way round.If you solve a problem correctly, it plays thenext move to check that you can follow upyou solution correctly; and if you give awrong answer, it shows you a refutation. Ithas the facility to update its collection ofproblems directly from some place on theinternet – I don’t know how often this placeis likely to be updated.A new feature is that MFoG11 can be usedas a ‘Go Client’, to connect to IGS, NNGS,and other compatible Go Servers. It is notthe best such client, but it does work, andyou may prefer to use it as your Go Clientrather than go to the trouble of downloadingand familiarising yourself with a differentprogram. If you do use it as a client, you canuse it to cheat, by asking its opinion of thestatus of your groups, the balance ofterritory, etc., as you play.You can use MFoG to play go using yourmodem with a direct telephone connection toanyone who owns any version of MFoG, orNemesis or certain other programs thatsupport the same protocol (the Go ModemProtocol).MFoG11 has a feature which I know of inno other program, and which I particularlylike. This is a ‘game score graph’. For anygame, you can ask it to display a graph of

how much White is winning by after eachmove. In general, this rises when Whitemoves and falls when Black moves; and itrises and falls by more than usual for whatMFoG considers to be particularly effectivemoves.

SummaryMany Faces version 11 has in one packagealmost all of the features available in anyother Go program. It is suitable as an intro-duction to Go for complete beginners, andhas plenty to offer to experienced players. I would not hesitate to recommend it toanyone who has a 32-bit Windows system(Windows 95 or better). It is attractivelypackaged, and would make a suitable presentfor a games player.

AvailabilityMany Faces of Go version 11 is availablefrom Schaak en Gowinkel het Paard for 110Euros. If you already own version 10, youcan upgrade it to version 11 for 55 Euros.

Friends of GoFriends of Go is a fund set up as part ofthe BGA accounts for donations frommembers to support promotion of Goin the UK. The fund is especially usedfor promotion amongst children. Atypical use is the purchase of cardboard9 x 9 sets to give to beginners. Anydonations should be sent to the BritishGo Association, care of the Treasurer.

Tony Atkins

IN THE DARK?

Reading out of order – 2In the previous article, westarted to look at how youcan sometimes save yourselfa lot of reading in life anddeath problems if youconsider the moves in adifferent order than the onein which they would beplayed. In this article, we’llexpand on that idea.Let’s quickly recap thestory so far. If it’s Black,say, to play in a life anddeath situation, he canfigure that the first movesare going to be numberedBlack 1, White 2, Black 3.Then, instead of searchingall the candidate Black 1’sexhaustively, he can startby considering which White2’s he is most afraid of. For each White 2, he thenlooks for a suitable Black 1and Black 3. If, for someWhite 2, he can’t find Black1 and 3, then he can’t affordto let White play it, so hemust start there himself, and no other first movesneed be considered. Wecalled the point of White 2a ‘critical point’.Many of us hope to avoidunnecessary reading byreplacing it with higher-levelreasoning of that kind, butyou’ll soon find that manylife and death problemsdon’t have critical points.Now we’re going to extendthe method of reasoning to

try to cope with some moreproblems.Problem 1: Black to play.

Are there any critical points?If not, how do you plan yourreading, or do you just haveto slog through every possi-bility?Diagram 1a: White has aneye on the 1–1 point andneeds another on the 2–2

point. White 2 here lookslike a key point, but it isn’t acritical point since themarked Black stones makethe 2–2 point a false eye.Diagram 1b: Well, whatabout the point that madethe eye false, then?

No, not critical. The markedBlack stones kill.

Diagram 1c: In fact, problem1 has no critical point. Butwe can still note that, ifWhite gets to play both 2

and 4 here then Black has nomoves 1, 3 and 5 to kill it.So at least one of Black’sfirst two moves had betterbe on one of these points.Once we know this much,we have to knuckle down tosome reading, but it’s easiernow, because we know whatwe’re looking for. Whichcandidate moves do we planto consider? Obviously, thepoints of 2 and 4 in diagram1c are two of them. We shalltry those first. If they don’tyield a solution, we can alsolook for other first movesthat take aim at these twovital points.Let’s walk through thereading, to illustrate theprocedure and also to warnourselves about how it cango wrong if we’re careless.Diagram 1d: Black 1 is thepoint that makes an eye onthe 2-2 point false. ButWhite 2 lives by making Aand B White’s miai.(Diagram 1a helps us to spotWhite 2 quickly).

WHAT THE BOOKS DON’T TELL YOU ~ PART VIIISimon Goss simon@gosoft.demon.co.uk

Problem 1❏ 1c

❏ 1b

❏ 1a

Diagram 1e: Next, we trythis Black 1. Now Whitemay try to engulf it with 2.This doesn’t look tremen-

dously promising, so let’ssuspend looking at it for amoment and see if there’sanything clearer.Diagram 1f: Black 1prepares to take away theeye-making point safely. Butwhen White plays it himself,Black needs to play at A to

make the eye false, and hecan’t, since White wouldthen capture it.Diagram 1g: Black 1 aims tomake sure the eye will befalse. If White tries to stopthat with 2, Black 3 works.

After White 4, Black 5 isatari against the two stones.After White captures Black3, Black pulls back at A andWhite is dead.Diagram 1g is lookinggood, but let’s not just banga stone down and expect anearly resignation. Thatwould be a classic case of‘katte yomi’. (Rememberthat concept, which RichardHunter described in thisJournal a couple of yearsago? It means sloppyreading, especially the kindthat arises from expectingthe opponent to play all the‘obvious’ moves.)A more careful look atBlack 1 in diagram 1g mayreveal that White can reply,not at 2, but immediately at4, and some further readingshould convince you thatnothing works for Blackthen. Diagram 1g isn’t thesolution.When we said that diagram1e seemed unpromising anddecided to leave it till later,that was fine. Unpromisingmoves fail quite often, afterall, and it’s just commonsense to look at the mostlikely chances first. Thatisn’t ‘katte yomi’, becausewe’re always ready to takea second look at theunlikely ones if we need to.It seems that now is thetime to take another look atdiagram 1e. Actually, it’seasier now, becausediagram 1g has shown ussomething.

Diagram 1h begins with thesame two moves as diagram1e, but shows how Black

can continue with 3 and 5 tomake a false eye. Next, ifWhite plays A, Black repliesat B, and if White playsanywhere else, Black playsat A himself. This is thecorrect solution. (To avoid‘katte yomi’, please checkfor yourself that other Whitedefences fare no better).

Generalisation andProblemsIn the previous article, wesaw how Black could helphimself to solve a life anddeath problem by askinghimself not merely “whatmoves do I need to play?”but also “which move must Iprevent my opponent fromplaying?” In the example wejust analysed, he couldn’t dothat, but he could still makeprogress by consideringwhich two moves Whitemustn’t be allowed to playboth of.In theory, you can increasethe number of opponentmoves to be prevented toany number you like. If theopponent can live uncondi-tionally by playing fivespecific points, then at leastone of your first five moveshad better be on at least one❏ 1g

❏ 1e

❏ 1f

❏ 1d

B A2 1

❏ 1h

3B 2 1 4A 5

of those points. The onlything is that this may behard to discover reliably, andit still leaves a lot ofreading, so maybe it’s ratheracademic when the numbersget that high. I must admitthat, if I’ve ever seen aproblem that could beanalysed by identifying threeor more critical points, Inever realised it.This idea probably isn’tlimited to life and deathproblems, though it’s usedmost often with them. Thething about life and deathproblems is that the strategicobjectives are clear, sothere’s some hope of identi-fying the vital points. But,whatever strategic goalyou’re contemplating,whether it’s invading amoyo, or making youropponent heavy, or whateverelse, it may be useful toconsider which are the vitalpoints your opponent wouldhope to play to stop youdoing it, and contemplateeither playing those pointsor something else that wouldrender the opponent’s movesineffective. Food forthought, anyway.But in the followingproblems, we’ll stick to lifeand death, and the numberof opponent moves to beconsidered will always betwo. Because the subjectcalls for clear thinking, thistime I haven’t swindledyou by including a problemthat has no solution. All ofthem work.

Problems 2 and 3 are Blackto play and kill. Problem 4,which is quite a lot harderthan the others, is Black toplay and live.

Answers

Diagram 2a: If White playsboth the marked stones,Black cannot kill with threemoves in a row.Diagram 2b: Black 1 failswhen White connects at 2,leaving A and B miai inWhite’s favour.

Diagram 2c: Black 1 is theanswer (as 2–1 points often

are in corner L&D problems).If White 2, Black 3 makes A and B miai in his favour.Diagram 2d: Here’s anotherdefence. Black 3 makes miaiof 4 and 5. The shape afterBlack 5 arises often in life

and death problems in thecorner and is well worthremembering. It would beseki on a side or in thecentre, but in the corner,White is dead, because whenWhite loses his outsideliberties, A is the only insideliberty of the big White group.

Problem 2

Problem 3

Problem 4

❏ 2c

❏ 2a❏ 2d

24 3 A

❏ 2b

Diagram 3a: This littleproblem illustrates a coupleof features if our theme thatare worth remembering.

Firstly, it seems so simplethat you might expect thereto be a single critical point,if anything at all. A lookslike one, but if White gets toplay there Black can still killin two moves with B and C.In fact, this problem has nocritical point, as you maylike to verify.The conclusion I’d like todraw from this is that, if nopoint stands out clearly as acritical point, it may not beworth looking for one. Evenif one is there, which is veryfar from guaranteed, youmay be able to read theproblem out almost asquickly as verifying whetherone is there or not.Anyway, in this position it’sclear that if White plays bothA and B, the group in thecorner is alive. So we canfocus on points A and B andothers that aim at them.Diagram 3b: Black 1 failsafter White 2. Next, if BlackA, White captures; and if

Black anywhere else, Whiteplays A.

Diagram 3c: This Black 1 isanswered by White 2, andonce again, Black has noway to kill.Diagram 3d: Neither of theobviously interesting pointsworked for Black, but isthere a way to take aim atthem? The attacks in

diagrams 3b and 3c wereboth refuted by a Whitemove on the 3-3 point, somaybe it’s the vital point?But when Black tries it,White can play at 2, makingA and B miai in his favour.Black can capture threestones in gote, but he can’tget the lot.Diagram 3e is the answer. It makes A and B miai inBlack’s favour and also Cand D miai in Black’s

favour. The thing I likeabout this problem is thatthe tesuji isn’t one of the

obvious vital points, but amove that takes aim at everyvital point at once.Diagram 4a: If you try toread this problem out fromscratch, it’s quite hard, but astriking feature of theproblem diagram is thatWhite can capture threestones just by cutting at the2-2 point.This diagram shows theposition after he has madethat cut and then captured

(the marked white stones),leaving Black trying to livewith three consecutivemoves. As you see, he can’t– to live he needs to add allthe marked Black stones,and there are four of them.White will be able to capturethe three stones unless Blackcovers the cut with his firstmove. So either Black mustbegin with one of the threepoints that do this, or, if hewants to let White cut, hemust add a fourth stone tothe sacrifice when Whitedoes so. The latter seems anunlikely approach, so let’sconsider the more likely-looking moves that cover thecut first.Diagram 4b: Black 1 herelooks like the sort of shapefor making eyes with, but

❏ 3a

C AB ❏ 3c

❏ 3b

❏ 3d

❏ 3e

C1 AD B

❏ 4a

White can play at 2. We’vealready proved that Blackcan’t let White capture thethree stones, so Black 3 isforced, and then White 4kills. If Black A, White justconnects at B and Black canonly make a false eye to theright of A.

Diagram 4c: If Black 1 here,White 2 kills. A and B aremiai in favour of White, andso are C and D. If Blackconnects underneath…Diagram 4d: …White justconnects at 4, leaving A/B

and C/D still miai pairs inhis favour. Black is dead.Diagram 4e: Black 1 maylook slightly implausible,since it’s a funny shapedoesn’t gain any liberties,but it’s the right answer.

It makes the point of White2 a critical point for thecorner part of the group(check!), so White needs toplay there if he is to killeverything. Next, Black 3appears necessary, but White4 kills in a way similar towhat we’ve seen before.However, Black 3 was ablunder.Diagram 4f: Instead, Blackshould have played at 3here. This makes a real eyeon the side. After White

captures the corner stoneswith 4 and 6, Black playsback where the triangledstone was and makes an eyethere after all; a classic ishi-no-shita.Diagram 4g: There’s a lotmore in this beautiful

problem. Steve Baileydiscovered this variation,where White offers Blackthe chance to live in thecorner while sacrificing thethree stones on the side, andBlack declines with Black 3,trying for everything. NowWhite 4 is his only chanceof anything. Please checkyourself that White mustblock at 6 and answer 7 bytaking the ko. This leads to…Diagram 4h: Black nowplays 1 as an internal kothreat (Black 2 would be

worse because it wouldleave White an internal kothreat of his own on the 1–1point). We now have a directko for the life or death of allthe Black stones. Comparedto giving up three stones andliving in the corner, this kois so unfavourable for Blackthat he should only embarkon it in desperation, sodiagram 4f is normallyconsidered to be the answerto this problem. But the ko isthere, and it could conceiv-ably be right for Black insome game situations.I shall leave to you the taskof proving that Black can’timprove on these lines byplaying to sacrifice fourstones in the corner.

❏ 4b

1 A2 3 4

❏ 4c

A B C D1 2

❏ 4d

A B 3 C D

❏ 4e

2 31 4

❏ 4f

6 2 51 4 3

❏ 4g

4 3 71 2

❏ 4h

The first five IOM Go Congresses werebiennial, but as the previous one had been in1999, some of us aficionados were feelingwithdrawal symptoms. As for the last two,the chosen venue was the small southernresort of Port Erin, but as the CherryOrchard Hotel is no longer available weplayed in the Falcon’s Nest Hotel.There was the habitual attendance of about50, including the usual assortment of parents,spouses, partners and offspring that the IOMattracts, and a goodly sprinkling of youngercompetitors. The overseas attendance wasdisappointing, consisting of Josè and Miguel,a father and son from Lisbon; however theymade a lively contribution, and walked offwith a prize each (see elsewhere). As usual, the pattern of events separated outmorning and afternoon tournaments, withone-off evening events. So it was easy tosplit your time between playing and sight-seeing, and many people did that. It is a fineplace for walking, bird watching and so on,and for wet days (we only had a few showers)there are well presented museums of Manxculture. What you won’t find in Port Erin aregourmet restaurants and sophisticated enter-tainment. I was pleased to find a brass bandconcert on the Sunday evening, while theofficial entertainment was a quiz night.In previous years the rest day for excursionshad been either the first Sunday or the lastFriday; don’t ask, this is the Isle of Man.This year the excursion day was a sensibleWednesday, when a free coach trip to Peel orRamsey was on offer. I saw my first baskingshark on the way; they’re huge things, butthey only eat krill etc. The trip worked quitewell, except that the Ramsey people gotrather short shrift as far as time wasconcerned. I declared independence, and cameback much later on the scenic coastal tram.For the dedicated, there was a one-day die-hard tournament back in the Falcon’s Nest.

Organisation was a team effort carried outby almost all the members of the IOM GoSociety. I know that the smaller numbersmake for ease of administration, but Icouldn’t help contrasting the hassles of therecent European Congress in Zagreb with thequiet efficiency of the Manx Go players.Now here’s a suggestion. I wouldn’t want todiscourage anyone from attending either theEuropean or US Go Congresses. But someof us are having our doubts about St.Petersburg next year (rather a long way fromthe city, and a reputation for mosquitos) orHouston, Texas (temperatures of 105 degreesFahrenheit). There won’t be an IOMcongress as such, but a few people areconsidering going across to the islandanyway, booking into the same hotel for thesame week, and enjoying the island andsome informal Go in the evenings and onany wet days. Any takers?

SIXTH ISLE OF MAN GO CONGRESSFrancis Roads francis@jfroads.demon.co.uk

Ing Fund The Ing Chang-Ki Weichi EducationalFoundation provides grants to Americaand Europe to help promote Ing Rulesand Go among children. Part of thegrant is Ing sets and timers.The European grant covers the cost ofthe Ing Cup, European YouthChampionships and part of theEuropean Congress. Part of the grant isalso available to support projects indifferent countries. The BGA has hadvarious grants and in 2002 receivedsupport for the Hampshire SchoolsProject.

Tony Atkins

IN THE DARK?

SmartGo:BoardSmartGo:Board allows you to play throughand edit Go game records. It can handle thetwo most commonly used formats of gamerecord, SGF format (with extension .sgf or.mgt) and Ishi format (with extension .go),and can convert between them. Like othergood game record editors, it allows you toinsert and to edit variations, and comments.Like the best game record editors, it displaysa ‘tree’ of the moves, so that you can seewhere the variations are, and can jump toany positions in the main sequence of play,or in any variation. However, the othereditors that I am familiar with (Hibiscus andCGoban2) display the tree permanently, withall its nodes visible, so that you can alwaysscroll along it and jump to any node.SmartGo:Board displays the tree ‘closed’, asa single node. To use it, you have to open itout, by clicking on this node to cause thesecond one to appear, and then clicking onthat to open the third one, and so on. And assoon as you jump to a node, the tree disap-pears again, requiring you to repeat thistedious opening-out procedure if you want touse it again. So I found it quite awkward touse the game tree in SmartGo:Board.Its most useful-looking feature is in diagramcreation. You can tell it to produce a diagramshowing the moves from, say, 30 to 46, andit does it, correctly, with the ko list (andother moves played on already-occupiedpoints) written out beneath the diagram.However, there are two snags. You can’texport this diagram in any way that I canfind, all you can do with it is print it to yourprinter. Also, the ko list is not wrapped, itjust extends off the side of the paper so thatyou can only read the first three or somoves.I do not believe that SmartGo:Boardcompares favourably with competing

products, many of which are listed at:www.britgo.org/gopcres/gopcres1.html

Hibiscus, which costs US$20, is an excellentGo game record editor, and can also convertbetween SGF and Ishi formats. Or, if youdon’t need this conversion facility, you canuse Yago, which is free (Yago can read andwrite both formats, but cannot convertbetween them).

SmartGo:PlayerSmartGo:Player is a Go-playing program. It is integrated with SmartGo:Board, anduses the same interface.It does not play well. I played it on a 9 x 9board, with time limits of 30 minutes each,and gave it four handicap stones. Despite itsusing 26 of its minutes, it lost by 81 points.There are many Go-playing programs thatplay better, and faster, than this. Forexample, WinHonte (from www.jellyfish-ai.com/winhonte.htm) plays significantlybetter, and faster, and is free.There is an option for it to display its‘thought processes’ as it thinks about itsmove – what move it is evaluating, what itthinks is dead, where it thinks the territoryis, etc. This can be interesting, but it can alsobe distracting, if you are playing against it,so there is an option to switch it off.An interesting feature of SmartGo:Player isthat it incorporates code from another author.Anders Kierulf has made a licensingagreement with Thomas Wolf, allowing himto include Wolf’s GoTools life-and-deathengine in his program. You can use GoToolsat lie.math.brocku.ca/gotools/applet.html.GoTools is, I think, by far the most powerfullife-and-death analysis program there is. In the form at that URL, it is only effectivein a position which is totally sealed off fromany interaction with the rest of the game –the possibility that a threatened group might

SOFTWARE REVIEW ~ SMARTGO:BOARD AND SMARTGO:PLAYER

Nick Wedd nick@maproom.demon.co.uk

be able to connect out by walking through abamboo join is enough to stop it fromfinding a good move. However, the form ofGoTools that is incorporated inSmartGo:Player seems more flexible. It cansolve life-and-death situations which are nottotally sealed off. Unfortunately, unlikeWolf’s web version, it does not always givethe result in a way that is easy to understand.When I follow the procedure in the manual,it solves the problem, but does not displaythe answer – I end up with a node labelled‘Node’, and no indication of what the correctmove is.These are both programs for Windows, byAnders Kierulf. They are both availablefrom the SmartGo web site:www.smartgo.com. SmartGo:Board costsUS$29, SmartGo:Player costs US$59.

Canterbury was the location of the 36thEuropean Go Congress. Its major sponsorwas Hitachi. The sun shone for the wholefortnight and much fun was had with recep-tions, songs, music, dice, cards and a trip toLeeds Castle. There were as well many goevents. A UK record of 234 players playinggo on one day was set. A three-way tie forOpen Champion involving Zhang,Miyakawa and Matsutomo, was split givingthe last named Japanese businessman thetitle. Alexei Lazarev of Russia was theEuropean Champion; he also won the placein the Fujitsu Cup by being second in theGrand Prix to Guo Juan.63 players attended the Northern, withstudent Matthew Cocke being the winner.Des Cann won the event at the OpenUniversity in Milton Keynes (Francis Roadsgot a special prize for entering twice).Matthew Macfadyen won Shrewsbury forthe fifth time. The Schools Championship

was held at Stowe School. Their team wasrunner up to Furze Platt A (Sam and JoBeaton and Chris Dawson), also theLightning Champions. Another team eventsaw the British beat the Japanese in a tent inBattersea Park. The 106-player Wessex sawTony Atkins lose the final to T.Mark Hall,CLGC being the best team and Jo Beatonwinning the Fred Guyatt 13x13 Trophy. The Meijin match kicked off in Amsterdam,with live commentary at the new EuropeanGo and Cultural Centre. Kobayashi Koichibeat Otake Hideo by 4 games to 3. He wonthe last game by 1.5 points, following abrilliant yose move, to become honoraryMeijin. The Oza was a win for veteranFujisawa Shuko; he beat the Meijin to win at67 years. In Korea 17 year-old Lee (Yi)Chang-ho was winning no less than six titlesand one of the so-called WorldChampionships.

TEN YEARS AGO

Tony Atkins

!"#$%"#&&$&"'(!"#$)*+,-#$&"'(!"#$%"&'%"($)%"*+,))"-&."/01.2,3!"#$%&'(#)&&($#*+,#"-./01"+%#2&3

4&5#6*'(7*11&+5#*+,&%8"3#1/+,#$0&3%$

9+#:&+,&+;$#!"$%#<+,=<$%*)>/$8",#?@@?

!""!""!""!""!""!""!""!

..$)/0#*$&!*##!1'2,'2$345$6*!

(789: ABA#CD@E#@BBB;<= ABA#CD@E#FFGG

#><?@ 6HI'8"$$J'&10.$"3K"L'&1

M0"+#?A*1NE015#I&+,*ONP*%.3,*O

A/+1B3#)$'*,#*$/$&(#%+/1+!C

Q>>#1*R&3#'3",/%#'*3,$#*''"0%",

7DDEFBBGGGHIJ>J7:KKHJ8HLM!"!"!"!"!"!"!"!

Not all that strange, actually; I borrowedmy title from the song which appearselsewhere; my contribution to the usualsong night on the final Friday evening.Still, after two European Congresses atwhich there was no waiting at all for regis-tration, the two hour wait was a bit of asurprise. There seems to be a reluctanceamongst European host nations to learnfrom each others’ experience.The Croatian Igo Alliance had chosen a largesports hall as the venue. Only the strongestplayers were in a smaller room elsewhere;this also served as the venue for professionalteaching and simultaneous displays. In someways it is good to have everyone together,especially when it comes to announcements,but on the other hand the referee (ViktorBogdanov of Russia) made no attempt insuch circumstances to maintain silence. A small stall offered vegetarian eco-friendlyfood, which suited me, but wasn’teveryone’s taste. It is also the first time that I have seen a small moveable post officewithin the playing area.Accommodation was in two nearby hotels;quite well-appointed, but not overlyexpensive, as the Croatian Kuna is fairlycheap. As a perk we all had free access toZagreb’s public transport throughout thecongress, and that was quite a bonus. Notthat it’s expensive, but the Zagreb tramsystem is remarkably effective, and linkswell with the busses which serve thesuburbs. I saw no traffic jam during thefortnight.There were the usual excessively long (inmy opinion) time limits; 2 hours 30 minutesfor supposed 4-dans like myself. But westarted at the somewhat un-go-player-likehour of 9 am, so there was a good chance oftramming off the short distance to the citycentre after lunch. And there is plenty to see

there; as well as all the usual NationalGalleries of this and Museums of that,Zagreb is well supplied with parks andgardens, as well as by far the largestshopping mall I have seen in an ex-Communist country.The Main Tournament rounds started moreor less on time and ran reasonably smoothly.I wish I could say the same for the sideevents. (See the song for more informationabout one particular event.) As with theregistration, there are well establishedtechniques for running these events expedi-tiously, but for whatever reason theCroatians didn’t use them.A popular afternoon activity was sitting inone of the numerous cafes playing pits, liardice or just gossiping. On many days it wastoo hot to want to do much else. And youcould get stuck in a cafe for anotherreason; we had some of the most violentthunderstorms I have seen for years, thefive-seconds-and-you’re-soaked sort.The Wednesday excursions were either tothe National Park at Plitvice, or a castlecrawl; both were on offer on bothWednesdays. I can take or leave castles,(especially when, as it turned out, the triplasts 12 hours), but Plitvice was a delight. A strange bit of geology has fashionedsome unusually beautiful limestone forma-tions, lakes and waterfalls. Visitors have tokeep to specially constructed walkways,which can get a bit crowded, but whichprotect the rest of the park from beingspoilt. We were told that it was also to keepwildlife and visitors apart, and as the saidwildlife includes bears and wolves, maybethat’s not such a bad idea.There was the usual influx of extra visitorsfor the weekend tournament, but I found somuch to do in Zagreb that I preferred not toparticipate. And as I have said, I learnt the

STRANGE THINGS HAPPEN IN CROATIAFrancis Roads francis@jfroads.demon.co.uk

hard way not to play in the evening events,so there was plenty of chance to sampleCroatian cuisine. This seems to centre onsome exciting-looking meat dishes, whichdidn’t really suit a vegetarian like myself.Fortunately there is a strong Italianinfluence, especially with regard to restau-rants, so I didn’t starve. Most Croatiansspeak a second language, which is just aswell, as Croatian does not trip readily offthe tongue. But it is as likely to be Italianor German as English.European Congresses have traditionallyproduced tournament journals, in anythingfrom two to six issues during the fortnight. I have often wondered whether these wereworth the trouble of producing, and theCroatians clearly decided that they weren’t.

They have promised to make all the usualinformation which they would containavailable on the web. Maybe that is thepattern for the future.I am very glad I went to Zagreb. I enjoyedthe city, and I found the people very kindlyand friendly. It was hard to square thepresent atmosphere with the horrors of whatis now called locally, ‘the Homeland War’.The tournament was, well, adequate, not sobadly run as to spoil ones enjoyment, but notquite up to the standard of other recentEuropeans. Next year the Russians will havethe chance to prove that they can do better inSt. Petersburg. They’ll need to have theiradmin well sorted out, as a high attendancecan be expected.

This may be difficult but do not be fright-ened. Settle back in a comfortable armchairwith a warm cup of coffee and a little timeto contemplate; gradually new perspectiveswill open up.This article consists of numbered definitesteps. Each step is a step in itself. Each steprelates to the steps around it.Try to read each step in itself separately; andthen pause to consider the previous steps –and open yourself to what is to come. Forthe steps are, in the main, an upwardsjourney.When you come to the end there is aninstruction to re-read this article. This is avery important exercise. If you do not re-read, the whole point of the article maybe lost.

There are two levels of ideas mapped out:(Paras 6–33) First Level (i) The Board (ii)the Position (iii) the Play (iv) Dynamics (v)Perspective (vi) Space Gates (a) move flow(b) plan flow – then – (Paras 34–46) SecondLevel (vi) Space Gates (c) Space Flow.Each of these two levels can beunderstood/related to the other in mind. To do this re-read each part while thinkingof the other part.The First Part is material/nuts andbolts/objects that you can see and think offor yourself while actually playing the gamein real liife: Subjectivity.The Second Part is Potential and Virtual.Space Flow is an implied idea of possi-bilites, an objectivity: an ideal context tolearn to be happy with. Read on:

GO FLOWRoger Daniel

i The Board1 The Centre of the board is the Gateway to

the rest of the board. The Side of theboard is a Gateway to the corners and thecentre. The Corners open out to the sidesand the centre.

2 Occupation of a position allows you toflow outwards and also to block theway.for your opponent.

3 If your opponent has occupied one of thestar point areas then you, by inspection,can ascertain how your opponent's flowcan be reduced - by placing a stone in thepath of the direction in which there mightbe flow and which you wish to restrict.

4 But, these seventh line points, which arethe areas which block the star pointoutflowing, are also Gateways to the starpoint areas.

5 The difference between (a) the 4 – 4points, the 4th lines and the 10th lines and(b) the 7th lines lies in the magnitudes oftheir distances from other parts of theboard and the number of directions awaytowards other important areas.

6 These are the major gateways of thestructure of the board.

ii The Position7 When stones are placed on the board they

cannot travel. Play must flow aroundthem.

8 As they form structure and the localposition, the development of the gamemoves through the intervening spaces.

9 Often the game develops by accretion andcrystal like growth from the initial seedplays. But the edges, of the formations,are, the Gateways through which thealready existing structures can flowoutwards to each other.

l0 These for consideration are the StructuralSpatial Gateways in Go positions.

iii The Play11 Realisation, of the preceding, 1,eads us on

to consider the possibility of TemporalGateways lying in the dynamic implica-tions of possible developments at anygiven position.

12 Go stones are placed on the board, and, inplacing a stone you simultaneously excludeyour opponent from making that play.

13 In addition, in Go, there are other waysthat one's opponent can be excluded frommaking a play :- by controlling oraffecting effectiveness.

14 A Gateway in a wall is the only place inspace through which travel can followafter the wall has been built.

15 Sometimes you want to be on one side ofthe wall and some of the time you want tobe on the other side.

16 Sometimes you want your opponent to beon the other side of the wall and some ofthe time you want your opponent toremain on one side of the wall.

17 That is why you build or you do not builda wall. That is why you do put or do notput a gate in the wall.

IV Dynamics18 Our real concern in Go is Dynamic walls

and Dynamic Gateways.19 A Gate can be an opening. But a True gate

can be converted into a blocked pathway,through which return is impossible.

20 The concept of a gate can be widened toa pathway which can be blocked at anypoint along it.

21 The board, with the game of Go struc-turally can be likened to a group ofmountains with passes between them.

22 The potential of possible developmentplay lines and their accumulation on theboard can be seen as a constantlychanging set of topographies, with pass-ages that can be blocked and unblocked.

V Perspective23 Maybe after the above you are beginning

to see how small the Go board really is.Partly this is true, because, the board fillsup towards termination of play.

24 The Dynamics and their characterchanges in class during the movementthrough the phases of the opening,middlegame and ending.

25 Cruciality and Criticality are the consid-erations in a successful1 execution of ameaningful game plan.

26 For those who think otherwise - areminder - that gradual keeping pace withyour opponent implies that exact countingand positional evaluation, constantly andcontinuallyF is, in such a game plan, alsocrucial and critical.

VI Space Gates27 Move flow, Plan flow and Space flow are

three concepts in crucial and critical gateturnings.

(a) Move Flow28 Move flow lies in sequential execution of

moves.29 A sequence where territory or group

survival immediacy depend on correctorder.

(b) Plan Flow30 Plan flow implies some goals and sets of

subobjectives, conceived of, to bereached, in the course of achieving one'sgoal and fullfilling some purpose.

31 If, one might think, that, if one achievesone's goal, and, then one's situation isequivalent to having a won game, then itis a Game Plan.

32 But in a game as complex as Go thebalance of the sum total of magnitudes ofmistakes is the explanation for the resultof the game.

33 Cruciality and Criticality are steps toleading advantage.

(c) Space Flow34 So what of Space Flow ?35 Space is an idea and reality, of totality, of

everything being taken into account.36 In calculating the total number of games,

one counts all the possible moves.37 In an actual game, the potential starts off

in all possible games - and, as each moveis played the possibilities disappear. Thenumber of possible games is slowlyreduced - to - one game playedcompletely.

38 All the possibilities at any one timeconstitute a Space through which theactual game will flow like a streamthrough a land.

39 The space around a position can bedivided up into WIN and LOSSSubSpaces and also sequentiallyconnected win and loss subspaces.

40 In Go a move can be played almostindependently from the rest of the game.This is another reason for the density andconnectedness of the space.

41 In Go the complete set of all combina-tions is possible except for the little holescalled 'eyes', and, the Suicide rule holes,and, the Ko rule holes; which render thedensity of the space not absolutelycomplete.

42 Since in potential the rest is completelyconnected, by exhausting all conceivablesequences the potential can rightly becalled a Space.

43 Space allows you freedom of movement;multiple pathways to the same objective.So the amount of Win SpaceFlow allowsyou room for manoeuvre.

44 Every move you make alters the directionof the potential spaceflow.

45 A Win GamePlan depends on thedirection that the potential spaceflow isopening up into, in terms of substantialadvantage.

46 Please read, once more, from thebeginning.

Black: Irina PatriciaMihai Petre Bisca Romania

White: Natasha Regan Matthew CockeUK

Komi: 5.5

Figure 1 (1 – 60)Black 9 is aggressive, andshould not give a favourableresult. However, it is complexand strange things happen inpair Go. The fight goes well forWhite up to move 30, whichmay be slightly better as abamboo joint. For White 32,descending at 35 would havecaused more problems.However, capturing at 40 givesWhite a thick position. White 46 may be too small,since the left hand side is openat the side. Playing a move tolook after 42 and 44 is worth-while. White could considerblocking at A. White 48 shows fighting spirit.However, Black 53 is very big,and invading immediately withWhite 54 gives black a target toattack.

Figure 2 (61 – 131)White 64, which I played, isvery painful and loses a lot ofpoints. I wanted to cut next at Bbut even if this works the lossfrom playing 64 is too great.White is in difficulties in thesequence played. Black 73 linkstogether nicely. White makessome small gains up to Black

111213

141516

17181920 21

232425262728

43444546

585960

❏ Figure 1 (1 – 60)

EUROPEAN PAIR GO ~ UK VS ROMANIAComments by Matt Cocke

616263

76777879

102 103

104105

108 109

110111

112 113

118119

121122

124125

127128

❏ Figure 2 (61 – 131) 114, 120, 126 @ 102117, 123, 129 @ 111

91. This move gives an interestinglife and death problem – is itnecessary? I believe it is, and it is alarge move in its own right.Cutting with White 92 helps toclarify the situation: White isaiming for a large centre, at the costof letting Black get the upper side.An alternative to White 92 is aninvasion around 96. White 106 alsoasks Black a life and deathquestion. It is a cheeky attempt togain a few points in sente, andBlack is correct to ignore it. I (lazily) leave the correct answer to White 108 as an exercise for thereader. As a hint, Black 109 allowsWhite to seize the vital point at110. This gets White a ko, and thegame is up for grabs.

Figure 3 (132 – 200)Cutting off three stones with White132 is big, and White is back in thegame. Black tries to salvagesomething with the atari of 133. The following fight is difficult, andthe short time limits added to theentertainment. White 134 isprobably too solid. and may bebetter played near 141, or firstrescuing the stone in atari. Theblack group looks alive after 141.Black 143 is skillful, creating thepotential of a later move at 146.However, it weakens the centralblack group. Black 145 forcesWhite to protect against this, andso is aji-keshi (a move thatremoves potential). White can now aim at playing C. If Black protects his eye with D,White can play a peep at 172. Thisseems to leave Black without eyes. I leave it as an exercise to the reader(again!) as to why White could notplay this without Black 143.

4748 49

51 5253

596061

66 6768

76777879

838485

93949596

❏ Figure 3 (132 – 200)

10 1112

❏ Figure 4 (201 – 263)

The black sequence up to 161 is slightly dangerous.A move at 163 is called for. However, White misstheir last chance to attack with 162. White does notseem to have winning chances after this.

World Amateur61 countries took part in the World Amateur.It was held in the Japanese provincial city ofTakayama during the first week of June.Unbeaten winner was Li Fu of China. Hongof Korea was second with 7 wins. The groupon 6 wins was: Kikuchi of Japan, Ri ofNorth Korea, Victor Chow of South Africa,Cristian Pop of Romania, Yu of Taipei, Guoof Australia and Pierre Colmez of France.USA topped the 5-win group in 10th.Britain’s Matthew Cocke (5 dan) was 22ndbeating Slovakia, Switzerland, Indonesia,Lithuania and Austria, but losing toAustralia, Hong Kong and USA. He hadbeen 26th last time he played in 1997.Ireland’s John Gibson was 57th beatingEcuador and Indonesia.

European Go CongressThe 46th European Go Congress was heldfor two weeks at the end of July in Zagreb,capital of Croatia. Zoran Mutabzijawelcomed players from 26 Europeancountries together with some Americans,Koreans and the usual goodly number ofJapanese. Professionals came from Japan,Korea and Europe, including Yuki Shigeno,Hans Pietsch, Katsura, Mr Chun, Mr Parkand of course Saijo (though possibly for thelast time). About 300 took part in theChampionship tournament. At the end,Alexandr Dinerstein, the pro from Russia,was unbeaten to be European Champion.Second on 8/10 was Dmitrij Bogatskiy(Bogackij)(6 dan Ukraine). On 7/10 wereMikhail Galchenko (5 dan Ukraine),Svetlana Shikshina (6 dan Russia), KaiNaoyuki (6 dan Japan) and ChristophGerlach (6 dan Germany). Heading the 6wins group were Radek Nechanicky (6 danCzechia), Csaba Mero (6 dan Hungary),Andrej Cheburakhov (Tchebourakhov)(5 danRussia) and previous Champion Andrej

Kulkov (6 dan Russia). Child star IljaShikshin (5 dan) was 11th. Highlight ofT.Mark Hall’s tournament was a game onboard 7 against Ilja’s sister, the otherRussian pro, Svetlana Shikshina; he endedtop Brit at 26th with 6/10. Of the 10 UKplayers the only other player on half winswas Colin Adams (1 kyu). Winning 9/10 wasGabor Szabo (4 kyu Hungary) and winning8/10 were Michel Spode (3 kyu Germany),Michal Hola (3 kyu Slovakia) and JanMilian (5 kyu Slovakia). The Weekend Tournament was also won byDinerstein, with runner up Csaba Mero andRadek Nechanicky third. Just over 200 tookpart. Nechanicky beat Dinerstein in the finalof the Lightning. Piers Shepperson made thelast 16 of the Lightning; there were no othergood UK results in the side events, apartfrom Francis Roads winning the Liar Dice(organised by the BGA). Francis alsopresented his latest song, together with twofrom Colin Adams and many old favourites,at the BGA-run song party.

US GO CongressThe 18th US Go Congress was held atConcordia University in River Forest justwest of Chicago, Illinois, during the firstweek of August. The university buildings arepleasantly centred around a courtyard andnearby to the Oak Park homes of FrankLloyd Wright and Ernest Hemmingway. Alsothere was the windy city and the great lakenearby to explore on the congress day off,with of course the alternative of playinganother tournament for the die-hards. 21professionals attended including Nakayama,Redmond and the American based pros likeJanice Kim and Feng Yun. 227 players tookpart in the US Open. Jung Hoon Lee won it,ahead of Jie Li and Hosuk Yi. However Libeat Lee to win the North American IngCup. In the Masters Jujo Jiang beat Feng

WORLD GO NEWS

Yun, winning the third final game on time.Pair Go winners were Ted Ning and Wang-Yu Chen. There were of course lots of sideevents including several events for childrenmostly win by Jin Chen and Erik Lui.German Horst Sudhof won nearly all theself-paired event’s prizes; Sinikka Siivolafrom Finland won the Optimist prize. Theprizes in the Bob High Memorial Poetry andSong Competition nearly all went to BobFelice, his winning song being “Learning toPlay Go in the West”. The new Lasker awardfor distinguished service to the gocommunity went to John Goodell who hastaught go for 45 years, including to RichardNixon.

Francis Roads and Mike Nash – the London team – playing at the European GoCongress in Zagreb

Photo:AJAx

British Pair Go PromotionPartnersThe British PGPP is an extension of thePGPP scheme run from Japan. As wellas agreeing to the aims of Pair Go, adonation of £50 is given to support thecosts of running the British Pair Goevent in a pleasant hotel environmentand gives a free entry to it. The fundwas swollen in 2000 by a donationfrom the Japan Pair Go Association.

Tony Atkins

IN THE DARK?

bgj128

Documents

Who Killed God

Chapter-01

The Dutch Republic In International Trade

(Tesla) - The Tesla Magnetic Car Engine

1 시스템분석입문

Puntuación PAEF,s

Improve the Color Quality Of Your Monitor

Tragic Heroes

Minne hävisivät soneran rahat

The Best American Humorous Short Stories

Steve Jobs' Commencement Speech at Stanford

Bodybuilding - The Rock Hard Challenge (Month 1 Training)

United States Constitution

Chapter 24

Venture Capital

Jan Van Eyck and the Man In A Red Turban

Iron Mills Essay

Cakes Recipes

расписание электричек Москва-Железка

Star Wars Trivia!