I’m another antipodean newcomer to this cryptic crossword game. The Guardian is the only one published in my local newspaper, and I’ve been attempting it for a few months now. I stumbled across the on-line access to it and also this excellent site a while ago. Many thanks to all of you for your blogs and comments. I think I’m starting to get the hang of it.

With regards to the grid, I only recently even noticed that it was always symmetrical. D’oh! I’ve just done a bit of googling and it appears that it has traditionally always been done this way. I’m interested in you more experienced solvers’ opinions about that. My (impertinent) newcomer’s opinion is that it would be much better to give the compiler the opportunity to tweak the grid as necessary, if it could improve the quality of the clueing.

Thanks again to everyone here. Now off to have a crack at Philistine’s 25684. Cheers.

My own observation about checking letters is irrelevant. If we start by blocking alternate spaces round the edge of a grid then only a maximum of 49 checking letters is possible and in reality the best I’ve seen is 48. Last week’s Rufus’ use of only 44 had no effect on the number of links.

I quite often find with all sorts of grids that NE and SW, say, are very incomplete but have become separated despite adequate links from NE and SW.

I do find the four mini-puzzle phenomenon irritating but I try to look on it as a challenge.

rho
Thanks for doing that tedious analysis. I could (should) have done it myself. Anyway it shows that my original comment was just a bit of nonsense.

Trailman
That all looks extremely complicated, it is not one of my fetishes to cut up crossword grids.(although it was the sort of idea which I used in my previous incarnation!)
I prefer the simple fact that solving 13 and 14 down undoubtedly makes 12ac much easier. Ditto for 5 and 7 down and 19ac.

Regarding checking, it is obviously more helfpul to the solver if more rather than fewer letters of solutions are checked. I don’t really mind either way, as long as the setter gives some consideration to the checked letters and how much of a clue they might give to the solver. -A-E- in a word is not terribly helpful, but if the clue for such is done neatly and without too much deviousness, it’s forgiveable (as in 26a, Rugus 25677).

I thought this good fun for a Monday and (Philistine that I am) can never seem to get too worked up about the grid. Didn’t know a Dutch painter named Hooch or this meaning of Attic. Must get around to buying a Chambers!

Great blog and for some reason really liked 14d

That said, I can imagine that grids that cry out loud for a Nina (like today’s) might be seen as less user-friendly than grids in which we have more first letters checked.
And indeed, grids that look like four separate puzzles are perhaps less stimulating too.

While I cannot be bothered too much, I must admit also that the grid used, for example, by Rufus in 25,659 (11 June 12) is not one of my favourites. If you can’t make too much of the cross in the middle, the puzzle almost falls apart. The first time I was really aware of this grid was in a Pasquale, who was so friendly to include some new (others would say obscure) words to make things even harder.
I forgave Pasquale, but perhaps not The Grid.
Generally speaking, though, I’m going with the flow.

If editors really like the figure out which grids we, the solvers, like or dislike, I would say: put these (or at least a selection) on the website of The Guardian and/or The Independent. Then ask people to vote: ‘yes/no’ or perhaps ‘yes/neutral/no’ – don’t make it more complicated (like using a 5-point scale). See what happens next.

To change the subject, normally I do not comment very much to Rufus puzzles. As a regular blogger of his alter ego I know what to expect, what I like and sometimes dislike and, most important, what I learnt since I started doing crosswords.
Today, however, I was a bit surprised by 1d (HOOCH).
Looking for a Dutch painter, having ????H, I was really puzzled – never heard of any. After I saw that it had to be Hooch, I googled a bit to find out that he was there!
Mainly English sites told me it was a painter, one that we, Ze Dutch, know as Pieter de Hoogh, with a g. Never heard anyone in Holland call him Hooch with a c.
Moreover, ‘de’ (lower case) is part of the surname, just like ‘van den’ in mine. And although it is, in an encyclopedia one would indeed list him (us! ) under H.
It felt a bit like saying ‘Goch’ instead of ‘ van Gogh’. Bit odd, isn’t it?

RCWhiting, in a comment on Enigmatist’s 25680, observed that of the checked letters 34% were vowels, and wondered whether that is high or low. I investigated this and my prelimanary findings are that the value in question is normally in the range 27-41%, with a mean and a median of 34%! I don’t know what RCW will make of that.

Now I don’t want to get into a lot of detail about sample size, statistical significance, deviation etc. Suffice to say that the only anomaly was in last week’s Rufus, where the figure was 48%.

I realize that this has to do with the content rather than the structure of the grid but I had to calculate the number of checked letters and these ranged from 52 to 60, with a mean of 56. In the Rufus there were 44. In percentage terms 34% to 37% norm, compared with 30% Rufus.

Today’s Rufus is 56 and 37%.

As far as sectors and connectors goes that’s getting complicated but it is obviously related to the simple number of checked letters.