Jump to content


Photo

Capitalization turned OFF is OFF


  • Please log in to reply
3 replies to this topic

#1 q256

q256

    Member

  • Members
  • Pip
  • 4 posts

Posted 30 April 2004 - 03:34 PM

:angry:

When I tell Spell Checker to ONLY check spelling (disable everything else) - I want that - nothing more - nothing less.

I can disable capitalization checking - but - it does not go away.

THIS and This and THIs should all be the SAME if I tell it so - but - it isn't.

Not a hard fix and it should be easy to do.

#2 Evan Gross

Evan Gross

    Administrator

  • Admin
  • PipPipPip
  • 2,991 posts
  • Gender:Male
  • Location:Toronto, Canada
  • Interests:This Place!

Posted 30 April 2004 - 07:05 PM

:angry:

When I tell Spell Checker to ONLY check spelling (disable everything else) - I want that - nothing more - nothing less.

I can disable capitalization checking - but - it does not go away.

THIS and This and THIs should all be the SAME if I tell it so - but - it isn't.

Not a hard fix and it should be easy to do.

The problem here is that "THIs" (mixed-case) is considered a spelling error, hence it gets flagged.

THIS and This and this (uppercase, proper case, lowercase) are valid case forms, and can be ignored by ignoring capitalization errors and words that are all uppercase.

The mixed-case flagging may be an issue. How often do you type words like this? This is the first request I've ever received for all mixed-case words to be considered correct...

#3 q256

q256

    Member

  • Members
  • Pip
  • 4 posts

Posted 30 April 2004 - 11:37 PM

Good example(s):

Washington washington < the second one is always a spelling error even though it is correct spelling, just not capitalized - I have turned off 'questionable' capitalization thus I expect only spelling error to be checked which is not the case.

Many of the compuer terms are either case depending on what you are doing with it. For example - xml and XML should be acceptable - another example.

Please also note :

example(s) < the '(s' makes the word example error every time also.

And/or < but it ignores a '/' so example/s works.

AND programmers DO type words like this. Typing in search parameters would be a case also when why capitalize when the systems are smart to only compare the words. So when you speak / write / communicate programming, coding or even with the xml java unix scripting errors - errors - errors.

I guess than I would ask what does disabling 'questionable' capitalization do ?

#4 Evan Gross

Evan Gross

    Administrator

  • Admin
  • PipPipPip
  • 2,991 posts
  • Gender:Male
  • Location:Toronto, Canada
  • Interests:This Place!

Posted 01 May 2004 - 12:47 AM

Good example(s):

Washington washington < the second one is always a spelling error even though it is correct spelling, just not capitalized - I have turned off 'questionable' capitalization thus I expect only spelling error to be checked which is not the case.

Many of the compuer terms are either case depending on what you are doing with it.  For example - xml and XML should be acceptable - another example.

Please also note :

example(s) < the '(s' makes the word example error every time also.

And/or < but it ignores a '/' so example/s works.

AND programmers DO type words like this.  Typing in search parameters would be a case also when why capitalize when the systems are smart to only compare the words.  So when you speak / write / communicate programming, coding or even with the xml java unix scripting errors - errors - errors.

I guess than I would ask what does disabling 'questionable' capitalization do ?

Right now, if a questionable capitalization is found, but you're not looking for those types of errors, it's marked as a spelling error. This obviously isn't the behavior you are after, but might be for others.

The correct solution is to have an option to ignore case differences entirely. This is often language-dependent, but the spelling engine we use has a way to do this. This functionality will exist in version 10.1.2, but won't be exposed in the UI (for various reasons related to Dutch language rules).

The other issues are ones I know about and will be getting to. You can partially solve the slash one yourself by adding a string value with the key "SCExtraWordSeparatorChars" containing the various characters you want to be considered as separators. Start with the string "<>=" and add more characters as you see fit. HOWEVER, this only currently works during Check Selection.

If you want a particular word to be recognized either lowercase, proper, or all uppercase, add it to a Learned Words reference as all lowercase.

Remember, this is a spelling checker for "normal" writing - not for programming. The engine we use is geared towards that, and has quite a bit of logic for distinguishing common, proper, and acronym case in all the different languages it handles. Same goes for abbreviations with or without ending periods, and other mixed-case special forms like "Ph.D.".

This is probably not well suited for programming terms, and it's often debatable what the correct case is for certain other computer terms (like XML, which is more generally considered to be correct all uppercase than lowercase).

I'll consider what you're after, but the correct solution must be one that also works for regular business correspondence (where case differences should not be ignored) and that works in all possible languages. That solution is not always trivial or obvious at first glance.