(Original post by ch_suginami-san: https://note.com/ch_suginami/n/n1f0d9ff49b31)
Although it's been a little while, I'll finally drop another article about lishogi. Just a few days ago I wrote a similar article, but it vanished due to "synchronisation" or something. But since I've noticed a few other things since then, I'll add them in as I go along.
Although globally the number of games played is slowly increasing, the total number is still rather small. There are still times when matchmaking is slow and it's somewhat difficult to find active users; but conversely this is evidence that there are still users who are excited to play on lishogi. I would count myself among these (whom I believe to be) Japanese people who are immersed in, and even contribute to lishogi. Contributing to the spread of lishogi within Japanese circles is also karakoro-san, whom I frequently see around and who also assists with the development of lishogi. I digress here, but they're stronger than me at shogi, and my challenges more often than not end in tears for me -- that's one of my impressions of karakoro-san (lol).
There are episodes like that, but to be honest, I feel like nothing will grow if we keep to ourselves. Little by little, it's fine if it's trivial, but I want to keep helping it grow.
The developers have also said in their blog posts that lishogi is still in its infancy, and there is still a beta feel to it. Even so, I would at least like to see a userbase that can get excited for it. Buzz generated by streamers is also something I want to see. Rather extreme shogi variants are possible and are quite interesting. Yep, something like this.
(YouTube video of Ayase Aya playing ROOK VS BISHOP shogi with viewers on lishogi)
An important area.
- The bug where uchifu-dzume (mate-with-a-dropped-pawn) results in a win has been fixed.
- To handle jishogi, the try rule has been implemented. The implementation of impasse declaration is being prepared.
- It is now possible to set the pieces in hand in the board position editor. This was a much-desired functionality. Furthermore, it is now possible to directly edit the board position when creating study chapters. SFEN format is support continues, so we should see more functionality added in the future.
These are topics that the above-mentioned karakoro-san has covered in detail in their posts, but I would also like to put forth my own views. There are a lot of things to look forward to for lishogi, and actually there are workarounds for some of these even now. I actually wanted to write about more than this, but since I can't recall everything now, I'll just write what I can remember.
The original lichess has this feature, but personally I think there are pros and cons. This is because the lichess version has puzzles to win material, or to checkmate. But these puzzles are generated from games between strong players, and then presented to the user based on their puzzle solving rating. The issue then becomes one of quality.
Since the problems are machine-generated, and the answers follow the actual game moves (TN note: they don't always; sometimes it's a continuation that was missed), unexpected solutions and incomprehensible moves may appear. On the other hand, a large number of problems can be generated and solved in a short span of time. Any lichess users reading now can probably solve about 50 problems in the time it takes to finish reading this post. However, I was also somehow dissatisfied, so I started to create problems from my own games. During this time, I realised that I know of very few sites which collect next-move problems, be it for chess, shogi, backgammon or other board games; especially so when considering Japanese sites (of course, there is plenty in general that I simply don't know about).
So I started a next-move problem series here on note.com. In this case, between creating the problem and the analysis by hand, a lot of manual work goes into each problem. That said, I do get the chance to reexamine the move I thought of playing, and leveraging AI analysis lets me understand what other good moves were possible. Thus I hold on to the belief that through further reflection and tackling problematic areas, if we can derive practical and correct feedback, it will help us improve. But it might only work for people above a certain skill level.
In the first place, if we focus on raising kyu players to a certain level, it's thought that solving many problems -- even machine-generated ones -- is a good gauge of strength, and also helps to build confidence. If so, then as a technical problem, collecting as many kifus as possible and creating the software is an urgent task.
Actually this is already filed as an issue in GitHub: an 81Dojo-like thorough postgame analysis mode is not available in lishogi. However, by creating a study after the game and changing just a few settings, you can get something more or less equivalent to 81Dojo's postgame analysis mode (including the ability to allow spectators, to keep everyone synced on the same position, the right to leave comments, etc.) And then -- although I think this is more for streamers -- being based on lichess code, lishogi can display large and clear arrows and circles on the board, and in various colours at that. Combining keyboard and mouse, four types of colours are available, and moreover it is simple to draw on the board -- a perfect feature for streamers (in making this available, the original lichess probably also had this in mind.)
(TN note: RMB, or shift + LMB = green; shift + RMB = red; alt + RMB, or alt + shift + LMB = blue; alt + shift + RMB = yellow.)
It'd be a waste not to use this feature. Now that you know how, if you do use it, let's do so to its fullest extent.
lishogi has an absolute selling point of being free forever, with no ads. I think this strongly reflects the overseas, and particularly programming, philosophy: beneficial assets should be shared. For better or for worse, since lishogi does not have the backing of the Japan Shogi Association (well, it is a recently created site after all), as long as matters of interest are carefully handled, lishogi probably will have the freedom to operate.
One area of operation would be the kifu database of online matches played on lishogi. The original lichess site accumulated millions of games amounting to tens of gigabytes' worth of text data every month in 2020. Now this can't be directly compared to storing shogi kifus, and even if lishogi gained more popularity, it still probably won't reach lichess' numbers. Nevertheless, even if we assume 10,000 games are played each month, making these complete kifu records freely available would definitely see them be used. Compared to other sites which mostly allow for game searches by username, the centralised organisation of all games on lishogi is without doubt a strength. Currently the focus on site development and the unfinalised format mean that this feature is not available yet, but it is certainly something to watch out for. And I am looking forward to the day this database eventually becomes lichess's number one strength.
Well, I described lishogi as an all-in-one shogi site, and there is another big reason for that. For example, let's say a person interested in shogi, particularly a young boy or girl, wants to learn the rules. Is there an online site that starts by explaining the basic rules, and lets them play online games when they are ready? Tell me if you know one, I don't know of any. I believe lishogi can break this barrier. That is, it teaches how the pieces move and the rules, everything gathered and in order -- and that whilst allowing you to freely move the pieces for yourself. And once the rules have all been learnt, you can build up confidence by playing against the AI, leading towards online games.
To get stronger, solve next-move problems as mentioned in point 1, and use the "Study" feature to do postgame analysis and to study specific strategies. I'm convinced that it is lishogi's absolute strength that all this can be consistently implemented. While I'm not sure that strong players from other sites will migrate to lishogi, it would be fine if players who become strong on lichess go forth to challenge those strong players on their home turf. When this happens lishogi will naturally gain a reputation, and good relations will be established.
What I would like to add here is that lishogi is currently host to a number of pretty strong players. Although ratings start at 1500, players around 1-dan strength can even fall quite some way below this initial rating. In fact, those who can hold a rating of 1500 are fairly strong. If you read this article and have confidence in your shogi, why not give lishogi a go? A wider world awaits you there.
The literal translation is "when you enter the village, obey the village", a Japanese proverb. The English equivalent is "When in Rome, do as the Romans do." (TN note: I switched the Japanese and English proverbs in the translation of the title and this first part.)
Why bring up this topic? Well, it's because of suddenly encountering a cultural difference. There is of course a certain cultural gap between Japan and overseas, and so even within the Discord server there are perplexing or heated discussions. At these times, the Japanese group will bring up "the Japanese cultural mindset", but this is not always easily understood. Well, that's to be expected; indeed one can only assimilate the local culture if you are right there in the locale. It is with this sense that I evoke the proverb in the title.
And even here, I thought the proverbs actually differ in a cultural sense, despite them sharing the same meaning. The English sense is "behave in accordance with your surroundings", while the Japanese sense is "obey". If I were to visit some foreign country, I would think the former to be easier. That is, "it suffices to match your surroundings." The Japanese proverb requires you to start by knowing the "rules to obey" in the first place. At this juncture it feels like the starting point is different. Whichever viewpoint is easier to adopt largely depends on the cultural background one grows up with.
So now let's return to the topic. The reason I bring this up is that the lishogi development team are not just concerned with shogi, but also listen to opinions while deeply respecting the Japanese culture. Were this perspective lacking, it would definitely have created problems, or would create problems down the line. Especially since there are many concepts difficult for even Japanese people to grasp, such as "the culture of smoothly reading the atmosphere" and "even though it's legal, it's not explicitly stated, and should be avoided out of concern for the customs or feelings of those around you." In the midst of these difficulties, as a Japanese person, I am frankly grateful for the constant attentiveness to Japanese culture and the spirit of respectfulness. Even at the best of times it is difficult to be a developer faced with the many requests of the users. So upholding just this one attitude, if there is anything that you are uncertain about, you can always ask the Japanese group for opinions (even though pings may end up flying my way.) I think there are many points that we modern Japanese can take home about this attitude.
Of everything mentioned so far, I think the last point is probably the least evident. But however many features are added, the project cannot advance without the trust needed to support them. I think this last point is the single most important principle for the mindset of the lishogi developers and users.
As the number of people increases, the number of unexpected problems will increase too. When that happens, we cannot but take a resolute stance. As long as development continues while respecting others and their cultural backgrounds to stand firm on this one principle, there is no way to not support this stance. Going forward, I will keep an eye on the future of lishogi, and do what is within my abilities to help.
Japanese translation! Yes, I've been working hard to translate lishogi into Japanese! Well, it's a bit of an exaggeration as there are other Japanese members moving things forward too. And I'm being counted on to translate official announcements and blog posts from the management. If a translation seems off, I will take the responsibility for it. Although let's just say, if you encounter such an instance, please help to translate too.
And there is a fascination with shogi variants verging on greed. There was a particularly lively discussion on Discord, but it seems like there are quite a number of variants that people want to see included.
Annan shogi seemed to be new to many, but doubutsu shogi, chu shogi, 4-player shogi... I glimpsed a "let's just put in everything that we can play" spirit. Out of those, chu shogi is particularly anticipated to distinguish lishogi from 81Dojo. That said, for maka dai dai shogi to rear its head... as someone used to modern shogi, that was rather incredible for me to hear. (Although as expected with such a board size, it's probably hard to implement.) I think it is important in the sense of cultural transfer for these variants, forgotten by or deemphasised by the Japanese people, to be in the spotlight overseas. If each game can survive independently, fun and new life will spring forth.
Although this was a long post, it's good to write something like this once in a while. If anything else pops up I'll write another post. It's also probably about time for me to write the next entry in the next-move series...
~translated from Japanese by Illion