I don't know what solvers mean when they say they had to guess.
I use what I call Edge Technique. Either mentally or using the circle, I place them where it would help to eliminate some squares. These are at edges. Then looking at the crossing clues, I can see if it works or doesn't. If it doesn't, I can eliminate the end square. Then keep working on it. This is also logic - finding where there cannot be positives.
I would think that a "guess" would be a random place where one would place a positive. This is not logic.
BB

Probably when users write they had to guess, they do not know how to use the advanced solving techniques, like the one described here. Every puzzle is solvable with basic logic. But on some puzzles it can take a lot of effort to find that single step that will move you along. Some people will give up on that and get along with the trial-and-error method or even just take a good guess about how to continue. I've even heard it called 'solving by intuition'.

There is a nice list in the Site News right now with a couple of puzzles that seem to require guessing for several users.

A nice challenge to solve those with pure logic only!

Jelga

As for me, I am starting to get quite annoyed with the '...but I had to guess' comments. Not the comment itself, of course there are solvers who don't know the advanced solving techniques, sometimes needed to solve a tricky puzzles, but the fact that most people who write that will not give the puzzle, nor the author the proper respect they deserve. I notice that the amount of low ratings on puzzles that have a lot of those comments tend to be a lot lower then on other puzzles without those comments. From that I deduct that users give ratings based on their own solving abilities, not on the quality of the puzzles!!

I have to admit that I still have some puzzles in my saved-list that I have been unable to solve without using logic alone, mostly 4 or more colored puzzles, but I am sure I will require the techniques to solving those as well someday.

I make an offer here for everyone...If anyone encounters a puzzle in which (s)he thinks guessing is required, send me a message, or post a screenshot in the 'get help' forum, I am 200% sure that me, or any of the other users can provide you with a logical way to continue the puzzle. Asking for help makes you only a better solver, meaning that you will enjoy the puzzles you try to solve a LOT more, believe me, solving a difficult puzzle with logic gives a lot more satisfaction than guessing...

Maarten

When the o's and x's don't enable me to prgress, I also use the "Edge Technique" described by you logiclover. It is sorta trial and error, and takes a lot of persistence, but it works.

I use the same technique you do around the edges of the puzzle, but I have always thought of that as guessing, because I am starting with the assumption (guess) that the filled square is in the corner, and then progressing from that 'guess' until I run in to an error (or complete the puzzle). If I do run into an error then I go back to the saved puzzle, eliminate the corner square and start over again with a 'guess' that the filled square will be in column 2.

Now you are telling me that is logic not guessing, so maybe other solvers who say they guessed are really using the same technique but don't consider it logic.

I agree that "trial and error" is a form of logic.

However, if you have to take many more steps before you find that your original assumption cannot be correct, it
definitely feels like guessing.

I suppose true guessing would just involve seeing if the picture looks right.

When I'm totaly stuck (as with some parts of "full of ghosts" that I'm working on right now) I pick one square, and from that i start solving using X to mark filled squares until I find an error, and then I mark the first square as background. I have always considered that guessing, but i guess it is some kind of logic behind it :-)

A comment on a puzzle by cantata: "I'm always amazed at people who think that since they needed to guess, guessing is necessary. Every puzzle is solved by the computer first. Computers don't guess."

It's not guessing. You choose the proper place to start with the assumption 'What if this clue began or ended here'.
I have watched LogicPict solve puzzles. It begins with the overlaps, for which there is an easy formula to use. That's why I don't like grids that are not 5-multiples. Then it begins to fill clues, usually at edges. If that won't fit, it deletes the attempt and marks a square with a negative.

You can do this a few rows or columns in, but must be careful to look at all adjacent clues. If you have not come to a negative after it begins to get "too much to keep track of", then clear it and start someplace else.

Bruce

I've often wondered if these could all be solved with what I would call "pure" logic. That is, there is always a square that can be completed with what is already known. Now it seems that the answer is no. The only way to complete many puzzles is to use "guessing," or "trial and error." I guess this can also be considered logic at some level, but in my book it's not "pure logic." Is this pretty much the gist of it? Thanks for any responses!

Jerry

Every puzzle in this site can be solved by what we call Single Line Solving (SLS), based on logic alone. When an author creates a new puzzle it is checked by a program for its solution. Only after it has been made solvable by SLS it is possible to submit the puzzle to have it published.

However, sometimes it is very hard indeed to find the next logic step. With these puzzles, trial and error or guessing will be used by a lot of people to continue. But if you really want to find it, the next step is always there! And if you can't find it, there is always the get help in solving forum. Post a screenshot with the puzzle at the point that you are stuck and ask for a SLS solution. There is always someone who will help you with it.

Have fun solving,

Jelga

Hello, I'm not a very good solver even if I've been visiting regularly this site for several years now  (didn't read the tutorials yet: I just discovered in another post they existed) and I've written this "had to guess" not a long time ago. For me, it was because I couldn't find any validation that my tries were good or not until a quite long time, and then proving me wrong. In desperation, I just guessed where the heck I was supposed to find a wing in this butterfly and then going on from this point, and it's something I find frustrating.

By the way, the picture was really gorgeous and my frustration didn't stop me to highly rate it.

I would say on the contrary. It's very logical. Mathematicans call this method "indirect proof". You can proove something by assuming the exact opposite and showing that it's nonsense.