- Blind mode tutorial
lichess.org
Donate

omg you changed it... AGAIN??

#61

@Dr_Zee it would help if you can send me the PGN for that. Chat or whatever.

@Dr_Zee it would help if you can send me the PGN for that. Chat or whatever.
#63

I must say as a long time lover of Lichess and user of studies (like, every day, for years) I do really, really dislike this change. Why have I got to click the plus button just to see the only comment at the end of a variation?

This seems deeply unpopular as far as I can tell and IMO I think this one is a miss and should be rolled back. Sucks for the devs who worked hard to implement it, but it's really just not an improvement at all and has made studies worse.

I must say as a long time lover of Lichess and user of studies (like, every day, for years) I do really, really dislike this change. Why have I got to click the plus button just to see the _only comment at the end of a variation_? This seems deeply unpopular as far as I can tell and IMO I think this one is a miss and should be rolled back. Sucks for the devs who worked hard to implement it, but it's really just not an improvement at all and has made studies worse.
#64

@Planet_CHESS465 said in #62:

So is the old one coming back?

I am the developer of the LiChess Tools browser extension. If I do it, it will "come back" when you install the extension...

@Planet_CHESS465 said in #62: > So is the old one coming back? I am the developer of the LiChess Tools browser extension. If I do it, it will "come back" when you install the extension...
#65

Another side point I would like to make is that maybe it would be good idea to examine what led to this change rolling out to production in the first place. Was this a requested feature? By whom? Alternatively, has Lichess reached the state of maturity whereby behaviour-breaking changes are being made without anyone really asking for them? A common phenomenon with UI teams in particular, I've found, and can often lead to unhappy users.

Another side point I would like to make is that maybe it would be good idea to examine what led to this change rolling out to production in the first place. Was this a requested feature? By whom? Alternatively, has Lichess reached the state of maturity whereby behaviour-breaking changes are being made without anyone really asking for them? A common phenomenon with UI teams in particular, I've found, and can often lead to unhappy users.
#66

OK, I looked into the problem and I finally understand what people are complaining about, and that's that the NEXT MOVE is the farthest in the tree from the current position.
Instead of
1 2a
\-2b
\-2c
one gets
1->2b
\-2c
2a
and that compounds as the tree branches further.

This has been brought on by the concept of "disclosure" in which one can easily collapse the variations and see only the "main line". One can see that programmatically it's so much easier to put 2b and 2c in a container and show/hide them as one, remaining with 1 2a in the second example compared with the first, but ease of implementation should not inform the UX of an application.

In the same time, these are structural changes, not stylistic ones, so simple CSS cannot solve your problem. Things have to be redrawn from scratch if one wants to get the previous behavior. Note that showing the first move in the first variation at the same level as the other two in the old style was relatively easy just with CSS styling. Now, that's impossible.

The concept of disclosure does have some merit. I can imagine situations where I just want to see the main line or the "main line" after some level of branching. It would help me to hide the noise. But personally I think that hiding the variations is not the goal, "collapsing" them is.

Consider a typical analysis. You start with the main move, but you are interested in some side variations, so you make the main move and the side variation first moves, so you know to come back to them, then you continue adding moves to the main line.

So you start with
1 2a
2b
2c
and then you start filling in the main line
1 2a 3 4 5
2b
2c
If the variations are too complex after filling them all, I would rather simplify to the PGN that I started with, than something else. In other words, you simplify
1 2a 3a 4a 5a
2b 3b 4b 5b
2c 3c 4c 5c
into
1 2a 3a 4a 5a
2b
2c
not hide the possibilities altogether into
1 2a 3a 4a 5a

This gets increasingly confusing when you navigate and have to pay attention to the "fork control" rather than the board or the move list to see what your options are.

And finally - phew long post! - the new rendering is inferior in terms of performance in the browser. Open a 3000 move study chapter and compare the performance between the old system and the new one.

I am afraid that this needs a proper redesign if we want to make this for the chess community.

OK, I looked into the problem and I finally understand what people are complaining about, and that's that the NEXT MOVE is the farthest in the tree from the current position. Instead of 1 2a \-2b \-2c one gets 1->2b \-2c 2a and that compounds as the tree branches further. This has been brought on by the concept of "disclosure" in which one can easily collapse the variations and see only the "main line". One can see that programmatically it's so much easier to put 2b and 2c in a container and show/hide them as one, remaining with 1 2a in the second example compared with the first, but ease of implementation should not inform the UX of an application. In the same time, these are structural changes, not stylistic ones, so simple CSS cannot solve your problem. Things have to be redrawn from scratch if one wants to get the previous behavior. Note that showing the first move in the first variation at the same level as the other two in the old style was relatively easy just with CSS styling. Now, that's impossible. The concept of disclosure does have some merit. I can imagine situations where I just want to see the main line or the "main line" after some level of branching. It would help me to hide the noise. But personally I think that hiding the variations is not the goal, "collapsing" them is. Consider a typical analysis. You start with the main move, but you are interested in some side variations, so you make the main move and the side variation first moves, so you know to come back to them, then you continue adding moves to the main line. So you start with 1 2a 2b 2c and then you start filling in the main line 1 2a 3 4 5 2b 2c If the variations are too complex after filling them all, I would rather simplify to the PGN that I started with, than something else. In other words, you simplify 1 2a 3a 4a 5a 2b 3b 4b 5b 2c 3c 4c 5c into 1 2a 3a 4a 5a 2b 2c not hide the possibilities altogether into 1 2a 3a 4a 5a This gets increasingly confusing when you navigate and have to pay attention to the "fork control" rather than the board or the move list to see what your options are. And finally - phew long post! - the new rendering is inferior in terms of performance in the browser. Open a 3000 move study chapter and compare the performance between the old system and the new one. I am afraid that this needs a proper redesign if we want to make this for the chess community.
#67

@TotalNoob69 said in #64:

I am the developer of the LiChess Tools browser extension. If I do it, it will "come back" when you install the extension...

Can you link the extension again please? Thanks.

@TotalNoob69 said in #64: > I am the developer of the LiChess Tools browser extension. If I do it, it will "come back" when you install the extension... Can you link the extension again please? Thanks.
#69

@lonelypeanut said in #19:

Old

New

PS: this imgur linking stuff is so annoyingly obtuse...

I also like the old way a lot, I don't want to lose my eyesight with these newfangled plus-minus chicken-crossed intersection thingies...

@lonelypeanut said in #19: > Old > > > > New > > > > PS: this imgur linking stuff is so annoyingly obtuse... I also like the old way a lot, I don't want to lose my eyesight with these newfangled plus-minus chicken-crossed intersection thingies...
#70

@LeafySnack said in #69:

I also like the old way a lot, I don't want to lose my eyesight with these newfangled plus-minus chicken-crossed intersection thingies...

The plus and minus things are

  1. Out of place
  2. Clunky
  3. Glitching and hard to press
  4. Obstruct the view of notations (especially in studies)
@LeafySnack said in #69: > I also like the old way a lot, I don't want to lose my eyesight with these newfangled plus-minus chicken-crossed intersection thingies... The plus and minus things are 1. Out of place 2. Clunky 3. Glitching and hard to press 4. Obstruct the view of notations (especially in studies)

This topic has been archived and can no longer be replied to.