In tech forms, when someone starts a new thread. It should be evaluated whether it has at least rudimentary sentence qualifiers, in the non-code tag fileds.
A crude list of requirements might be: at least on capital and lowercase letter, one of .,?!;, perhaps even no excessive usage of internet common acronyms(possibly to slow?).

Why should such a scanner be implemented:

  • "i don lik seein stuf lik thi in a threa"
  • May prevent some unreadable inquiries, being posted in the forums.
  • People attempting to post such poorly formatted post, might rethink their message and give it an overhaul.

Obviously, I am, as stated, talking about checking for the bare essential characteristics of a readable sentence. Otherwise anything more grammatically correct than that, would prevent a lot of people(including myself) from ever being able to post! lol
Although there could always be an "ignore it" option.

Oh, of course checking if there is more than 45% capital letters, which is probably a sign of using all caps or something like "tHiS".

We don't even have a code tag checker (which would be relatively straightforward to implement). What makes you think we'll invent a literacy checker any time soon?

commented: :) +12
commented: QFT! :-) +26

A literacy test would only work for people whose native language is English. Non-English speaking people might have a difficult time to meet the requirements.

Agree with ad.

Also lots of threads start with a log/error message/link instead so any sort of filter wont work well.

We can just get the code from Bable Fish and modify it to properly format the translations to be syntactically correct for the new language. Should be trivial... :icon_twisted:

Also lots of threads start with a log/error message/link instead so any sort of filter wont work well.

Why do you think that a log/error message won't make it past the filter ? OR are you thinking that they are themselves written in so bad a langauge/ grammar ? Now C'mon

OR are you thinking that they are themselves written in so bad a langauge/ grammar

Yes? a log has bad grammar

e.g

01 - BHO - SearchDefault "http://www.search.com"

would be an acceptable line in something like a Hijack This Log. However, to a filter it would look no different to:

Buy Shoes Today - "www.ispamuswithshittyshoeadverts.com"

01 - BHO - SearchDefault "http://www.search.com"

would be an acceptable line in something like a Hijack This Log. However, to a filter it would look no different to:

Buy Shoes Today - "www.ispamuswithshittyshoeadverts.com"

I don't get you.

Also we cannot be enforcing grammar (because I feel it would be too complex a task) but we could to some degree enforce proper punctuation marks, and I guess that's what the OP was mentioning.

Do you mean as in how Microsoft Word marks perfectly good grammar as bad, because it picks the wrong word to use as the verb?

I have never seen a grammar program that works right.

I have also never seen a spelling program that knows all of the words.

Get rid of this absurd idea.

commented: Exactly :) +36
commented: echo! +11

When did I ever say it should check grammar? Rather I said it should see if there was things such as: excessive capitals, no capitals, and other such things that might indicate a poor post. Checking grammar would be an exhaustive process.

I meant that it would perform just as poorly as the grammar and spelling checkers do. Someone with a perfectly valid post would be caught by this contraption, and would be unable to post what he means.

I want an email filter that recognizes zxnrbl in the email and throws it in the spam can. But that is just as hard to implement.