What went wrong with AlphaGo ?

[Javaness2]

Nothing much went wrong with the software itself, but what about the knock-on effect we were expecting. I suppose that most of us remember the Hikaru Boom(HNG), when all these kids started playing the game. Across the world there was a big surge of new players, and all associations saw a boost in numbers. AlphaGo brought a huge publicity kick, it also brought some money to organisations like the EGF. What happened that we were unable to harness that to bring in more players. The figures below are for UK membership. I would have probably expected it to be one of the organisations that would benefit most, since it had DeepMind based in its country, and it had advanced knowledge that it was going to expect a mammoth publicity boost. It seems to follow the trend of most countries though, just a gradual decrease post HNG...

Code: Select all

[Hide]1972 516
1973 586
1974 682
1975 1048
1976 984
1977 991
1978 865
1979 685
1980 703
1981 679
1982 630
1983 541
1984 523
1985 491
1986 507
1987 519
1988 515
1989 518
1990 538
1991 577
1992 604
1993 575
1994 562
1995 534
1996 536
1997 644
1998 619
1999 639
2000 659
2001 655
2002 690
2003 703
2004 701
2005 653
2006 596
2007 591
2008 492
2009 496
2010 487
2011 467
2012 459
2013 447
2014 440
2015 430
2016 457
2017 417[/Hide]

[tchan001]

HNG inspires people to keep progressing as they reach higher to find the hand of God.

Alphago showed people that even pros can't win against AI so it made many people wonder why they should even bother especially when people they play against online might be using AI to play against them.

[Kirby]

I also liked the cultural aspect of go, like people following pros in Asia.

Pessimistic rant:

These days, whenever I visit L19, I can expect some sort of post about how some joseki is wrong because of what Leela says, or perhaps some sort of stuff about whether Carlo cheated with a computer or not.

There are exceptions, I guess, but the overall feeling is different from before.

Personally, I'm not even interested in going to the US Go Congress this year, which is something I'd always been excited about since I started attending 5 years ago.

[Javaness2]

It's very probably true that HNG was more inspirational, and more inspirational to young impressionable minds, but I don't agree that AlphaGo was something that sounded a death knell in the ear's of Go players. Most of us already realised a long time ago that we were not going to become pro, that we were not, by playing, walking into some mystical transmogrification of utopia. I guess a few people were put off, but equally weren't a lot of people turned on? Computery types enjoy the new tools, geeks and nerds around the world visiting go sites because they were interested with the whole AlphaGo thing. How come they weren't, in some tiny degree, hooked into playing?

M.Lim wasn't the only one who described Go as a drug.

[Javaness2]

Pessimistic rant:

Nice rant, but that's a question of what went wrong with LifeIn19x19. Fewer people around, fewer people discussing the game. You're an admin, ignore the problem or call it out and pretend it is the fault of the users.

[dfan]

As another data point, I returned to go because of AlphaGo after a lifetime of checking in on it every once in a while. I'm still going strong two and a half years later and have been taking it seriously enough to play in the AYD for the last four seasons. I love being able to analyze games with tools such as Lizzie. I'm going to my first Go Congress in a few days and am pretty excited about it.

I don't know whether there are more dfans or more Kirbys out there. Ideally there would be a place for all of us.

[Gomoto]

Chill out people, it is just summer in some parts of this planet

[jlt]

According to Javaness2, UK membership peaked in 2003. For some reason, the European Go Database shows a peak for UK in 2008. The graph below shows that Russia managed to develop go well after that.

capture.png (33.71 KiB) Viewed 14656 times

[Gomoto]

I am so sad no politics in this forum.

I would have some hints for the decline in UK and incline in Russia

Dont bother. I will not cross the line

[Gomoto]

If you want to increase the number of players I would recommend ...

some forms of advertisement:


Start Go clubs in schools,

Show Go during public events,

Put up flyers for your Go club at the universities,

...

(And dont overestimate the AlphaGo effect. What was the effect of Deep Blue on Chess followership?)

[Uberdude]

According to Javaness2, UK membership peaked in 2003. For some reason, the European Go Database shows a peak for UK in 2008.

Probably he was counting BGA members whilst you are counting EGF ranked players from the UK. These two groups will have considerable overlap, but some will be in one and not the other.

[Uberdude]

Personally, I'm not even interested in going to the US Go Congress this year, which is something I'd always been excited about since I started attending 5 years ago.

That's a shame. Am I right in thinking this is largely because you have fallen out of love with go (in large part due to arrival of strong go bots) and not because you think, were you mentally in the same place as a few years ago, the congress will be worse now than before? Or does the fact Andy Liu and Ryan Li will play an AI pair go game, or during a lecture they might talk about AI inspired moves like early 3-3 invasions, or your opponents play such moves actively turn you off?

[eyecatcher]

I just restarted studying and playing Go after a long break. I have to admit that Alpha Go`s win`s against Lee Sedol and Fan Hui unconsciously demotivated me... although it took me some time to accept that.

But... actually it is not that significant for my love / appreciation of the game. Compare to other sports: running, lifting weight etc. the fact that a machine can climb a mountain much faster or can easily lift tons of weight... or speaking about more abstract things... can calculate complex math-tasks in milliseconds, does not make it less interesting when people training to lift weight or try to calculate without the help of a calculator.

The game itself never changed, it is still alternating black and white moves on the boards, there are only now some stronger players around.

But yes, coming back to the topic... Alpha Go`s win was generally perceived as a demystification of GO, bringing it down to hard facts, wich are able to be calculated. No more esoteric mystery here. Edit: That is not a big marketing clue: "hey we have something here wich was regarded as beyond computer knowledge, but it is not ... but it is cool!"

When i started with Go, i did liked to think of GO, as being complex like playing 4 or 5 Chess games on one big board, you have a lot tactical battles on the board going on, and at the same time you should make some strategical decisions how to let the battle-fields - kind of - work together.

Even while i like Go more, i like to play chess a lot. I have no problems with the fact, that computers are better at it.

To make a much to long story short:
I think i was sad when Alpha Go won because it proved that Go is like Chess, not a mysterious / esoteric game, and some unconscious part of me did like to think of GO as something mysterious/spiritual beyond pure logic.

But now, after i accept that GO is in fact nothing more than logic, i`m fine with it.

Ah, and here is a related interview:
"Ke Jie wants you to know weiqi is fun -- and easy to learn"
https://youtu.be/OCevCII1zo0

[Gomoto]

Thanks for the link to the likeable interview.

I for one welcome the proliferation of go wisdom by AI. I like the demystification of the local go bullies as well

[daal]

The interest in go seems to be steadily declining - why should AlphaGo change that? People who were interested in go before AG are probably mostly still interested -excepting those who have died or become senile, and those who were prompted by AG to give go a try were probably quickly discouraged. In these click and swipe times, with no hope of ever being the best, it's not surprising that so few want to embark on such an arduous journey. AlphaGo just makes go all the more daunting.