A curious spelling 'help'; other spelling encounters

  • I type “discusses”; it gets changed to “discusse’s” (then gets marked as an error when “Check spelling …” is checked in Corrections):
    The proximate cause for this is that I substitute “se’s” for “ses”. (The very abbreviated reason for this has to do with my addressing the third-person singular indefinite pronoun default in English in a way that my father initiated for his own use several decades ago.)
    The ‘Scrivener’ cause for this though is more fundamental: The speller’s substituter applies itself to letter strings that are part of a word. This is an unusual approach to user’s substitutions (I’ve not encountered it in other apps), but in itself, this does not describe the full scope of the technique. Of course, the ability to change ‘things’ even right up against or surrounded by letters is of use for symbols and punctuation, em-dashes, ‘smart’ quotes, sentence-initial letters, but …
    When I type the (English) nonsense word “dsesd”, I get “dse’sd”.
    Is this ‘expected behavior’?

‘Learning’ it doesn’t help: And I’d not want to start filling up Scrivener’s Learned Words (my Personal Word list) to facilitate the checker’s uroboros-like change process.
Whether Tools > Corrections “Check spelling as you type”, “Correct spelling errors as you type”, are checked or not, matters not. And though it’s nothing I’d have wanted, I’ve checked both my Personal Word and Substitutions lists for something close to “dse’sd”, even inadvertently added: Found nothing.

  • I have a broader question that might explain some other not-too-dissimilar speller reactions.
    What’s the sequence of spelling actions when, say, the options I have checked are …
    Check spelling as you type;
    Correct spelling errors as you type;
    Both Auto-Capitalizations;
    But NEITHER Word Auto-Completion;
    and all Substitutions checked except “Disable …”?

  • It seems the auto-corrector is somewhat insensitive to mistyped contractions … dont, Ive, …
    (Curiously, it sees a mid-sentence “ive” (standalone) as an error, which is part way ‘there’, contraction-spelling/correction wise. But if I type “ive” (standalone) at the beginning of a sentence, the speller auto-capitalizes ‘the first word’ of the sentence/line (as I want done) – then immediately, sees “Ive”, thinking it’s a name (?), and doesn’t see it, flag it, as an error.)

  • In the Substitutions popup, in its add (+) popup, the initial focus is on the OK button, regardless that I have not yet given it anything to ‘OK’. Would it not be better were that focus be on Replace?
    (It would be nice too were I able to tab/keyboard-get to the +/- buttons.)


And another instance of the spellchecker’s not liking ‘ses’ in words–‘responses’–which it changes to response’s as soon as I type the trailing space or punctuation, whether or not I have “check spelling as you type” checked. (Which checked I usually do.)

(I do not have “Correct spelling errors as you type” checked. And as has been true in similar cases, response’s is not in either my Personal Word or my Substitutions list.)

I address such occurrences by typing that trailing ‘character’, then back up and delete the apostrophe.

As noted earlier, this is due the speller’s applying substitutions to strings pre/appended to words or embedded: An unusual application.


Personally I never use the auto-correct feature and have it turned off. I find it’s a pain and more trouble than it’s worth.

I have used Windows Scrivener since it was in beta and I treat it as an advisement only, when the spelling feature marks words/usage as errors. I have found that I must “train” it up with the learn spelling (right click on word) option. This is less of an issue for me because I write F & SF and I was expecting spellcheck to throw a fit.

Once it was trained, it stopped flagging words.

I may be wrong about this, but I believe the developers are investigating a better word handling system for upgrades.

Change your “ses” → “se’s” substitution to " ses" → “se’s” (note the leading space on the trigger word)

That way it only fires when ses is a separate word.

Thank you both.

Thanks Talktidy,
Actually, your way is as I have my current (ie, pre-Scriv) software, ‘InfoSelect’, spell checking set. And for the sake of 20+ years’ worth of habit, I’d likely run Scrivener the same way – except that

  • IS’s manually-triggered spellcheck applies my shorthand substitutions, of which I have a few hundred and counting now during my transition effort. Were Scrivener to do this, it would have not only saved several transition days’ work, it would facilitate my note-taking ‘outside’ going forward and my importation of its results. (Certainly, this transition manual effort involved gives me the ‘opportunity’ :unamused: to cull my outworn shorthands and to insert SarsenLintel’s suggested delimiter spaces.)

There are always tradeoffs … .

And thank you SarsenLintel,
I have now inserted the ‘delimiter’ space before both my trigger ‘word’ and its substitution. (I’ve got to admit that in hindsight, the stone I was tripping over was pretty obvious :unamused: – Thank you very much for putting the light on it!)

But there is a limitation: Because there is no leading space at the beginning of a paragraph, the substitution doesn’t see a space, so does not act. (I know, picky, and rarely an issue: Just something to be aware of when I apply the technique to other substitutions. :slight_smile: )

There are always tradeoffs … .

Is this irrespective string-replacement practice common in spellcheckers? (As noted earlier, I’ve not encountered it before.)

(Just today, I saw an aerial shot of you … Stonehenge. :wink: )

Thanks again to both of you: I appreciate the attention. :smiley:

To my knowledge not many people pick up on that. Or maybe they do and just never say. :laughing:

First off, apologies for any errors - I’m almost entirely on the Mac these days and so am getting rusty on WinScriv.

I think it’s more just a case of you’ve set up a substitution to look for a specific set of characters and then replace those with something else - so it’s just doing what you’ve told it to.

Another option for the delimiter - instead of spaces pick a character you would never (or very rarely) use, that way your substitution will only fire when you specifically tell it to. I use \ takes a little bit of retraining but quickly becomes second nature.

If you’re going to have a lot of such substitutions you might want to look into a dedicated program for it (and that way they’d be available no matter what program you’re working in.) On the Mac I’m using TextExpander, and I believe they have a windows version in beta now. What I like about it is it keeps an eye on what you’re typing and if it sees you typing the same thing a lot it suggests “why don’t you make a shortcut for this?” And if you’ve got lots and you find yourself forgetting shortcuts it can remind you about those too.

I’m sure there’s lots of other similar programs too if TextExpander isn’t your cup of tea.

Yeh, I’ve known about you for decades :wink:

Thanks on the extraneous character idea. Actually, I do/did just that in InfoSelect. There, there is a nearly universal ‘x’ at the end of the trigger strings to ensure the red-squiggly eye-catcher before the manual check: Singular possessives then likewise nearly always have an ‘s’ after it.
(InfoSelect’s substitution process treats each string as a word: Discusses does not become discusse’s.)

‘ses’ is one of the relatively few * of my substitutions that doesn’t have the ‘x’, mostly because I use ‘se’ etc. in conversation with my spouse, siblings, their spouses and progeny. (They put up with me … on this. :unamused: )

*I’ve not finished counting/assembling my list of substitutions for manual input to Scrivener, but so far, it numbers about 300.

As an aside: The past couple of days I’ve been cleaning up the ~1,500 words I use that are apparently not in Scrivener’s dictionary: I’m about halfway through ‘spellchecking’ the list to get them onto the Personal list. (I could have pasted them into that list, but I am using the opportunity to trim the list a bit; and the effort triggers a few new words for the list.)

Thanks again.

(I don’t usually get back so quickly … just had some time tonight before bed.)