Nothing

[qwerty]

Just forget it. I never asked.

q.

[Eve]

If only English was that simple.

Yes, adding 's' is the first rule. But after that it's a long tail of additional rules and special cases with diminishing returns ending with a long list of special cases that apply to a single word.

Even if that was solved, it would only work for nouns. Verbs have a completely different set of rules for each tense and acronyms and names shouldn't accept any variations.

[Glassboy]

More accurately pluralisation in English follows the rules of the source language of the word, and even then sometimes there are no rules and it is arbitrary and based on common current usage (e.g. Tolkien invented elves and dwarves because he thought it looked better).

cow - cows [affixing s|es|ves]
ox - oxen [affixing en]
goose - geese [vowel modification]
tui -> tui [no plural in source language]

or irregular forms like person to people.  

It's something that even people regularly get wrong; it's very hard for a machine to get right. :-)

[RoyC]

There are whole libraries devoted to the intricacies and irregularities of the English language. When Glassboy talks about 'source language' you might take a half-hour or so to consider how many source languages there are for English, starting with all the linguistic groups that invaded the British Isles, followed by all the linguistic groups that got invaded by the British, adding in the partner/rival countries such as the Dutch, Spanish and Portuguese, and then considering all the franchise countries that work their own magic - the Kiwis, Australians, Americans, Canadians, South Africans etc. The wonder is that any of the native speakers manage English, let alone how so many 'second-language' speakers can speak and write it so well.
But, as Simon and Glassboy suggest, there would have to be a monumental set of rules, exceptions and alternatives defined to get a computer application to decipher it all.

[Paolo F Cantoni]

More accurately pluralisation in English follows the rules of the source language of the word, and even then sometimes there are no rules and it is arbitrary and based on common current usage (e.g. Tolkien invented elves and dwarves because he thought it looked better).

cow - cows [affixing s|es|ves]
ox - oxen [affixing en]
goose - geese [vowel modification]
tui -> tui [no plural in source language]

or irregular forms like person to people.  

It's something that even people regularly get wrong; it's very hard for a machine to get right. :-)

Indeed, but the design of the glossary is sub-optimal...  

It should have the base term (singular) plus fields for the plural and singural (a concept I have created - "cow(s)" is the singural form of cow, encompassing both the singular and plural forms).  The glossary can then be populated however you like (we use our Onto-Terminological Model (OTM) to define the terms and concepts).  But the highlighter can use all three forms to highlight the right word and provide the definition as the tooltip.

Our Onto-Terminological Model holds the three spelling against each term and the concept provides the definition.  We automatically maintain the glossary - unfortunately 3 entries per term from our OTM as our OTM is updated.  This also allows us to define compound terms correctly:  Sergeant at Arms, Sergeants at arms, Sergeant(s) at Arms.

Paolo

[Uffe]

And even more uncommon is pluralization of verbs

I am sorry, but we are not in agreement.

But seriously, while this would be a cool feature, the cost/benefit ratio is simply far too high. The glossary is, ultimately, a minor EA feature and it would be very hard to get a function like this up to an acceptable standard.

/Uffe

[Uffe]

Indeed, but the design of the glossary is sub-optimal...  

It should have the base term (singular) plus fields for the plural and singural (a concept I have created - "cow(s)" is the singural form of cow, encompassing both the singular and plural forms).  The glossary can then be populated however you like (we use our Onto-Terminological Model (OTM) to define the terms and concepts).  But the highlighter can use all three forms to highlight the right word and provide the definition as the tooltip.

Our Onto-Terminological Model holds the three spelling against each term and the concept provides the definition.  We automatically maintain the glossary - unfortunately 3 entries per term from our OTM as our OTM is updated.  This also allows us to define compound terms correctly:  Sergeant at Arms, Sergeants at arms, Sergeant(s) at Arms.

Paolo

This actually sounds very interesting. Not because I have a customer asking for it, but simply out of personal interest.

Is this proprietary or something you could communityize? communify? communalize? release?

/U

[Uffe]

, the cost/benefit ratio is simply far too high.

No. As I said the current highlighting is not done via a semantic analysis but by simple string comparison. So just add the "s follows string" rule to underline also the "s". Simple as that. Works in more than 90% of all cases I'd say at almost no cost.

q.

1) 90% - citation please.
2) You're assuming that 90% correctness is acceptable to most users. This is not usually the case when dealing with language issues.
3) The glossary is user-defined. You're not guaranteed that the terms are in English, and there's no way to specify which language it's in.
4) "Almost no cost" - weasel word. Unless you've actually worked on EA development, I don't think "how hard can it be" is an acceptable argument.

/U

[qwerty]

 

q.

[Paolo F Cantoni]

[[size=14]SNIP][/size]
This actually sounds very interesting. Not because I have a customer asking for it, but simply out of personal interest.

Is this proprietary or something you could communityize? communify? communalize? release?

/U

It's quite intimately tied up with our OTM, so extracting it might a bit difficult at present (I have no spare bandwidth).

However, the actual Glossary population process is quite straightforward.  The trick is to get the source of data.  The three forms are held as Tagged Values against the term, which itself has a tagged value indicating it's a glossary term (since we hold others).
The concepts are separate objects connected to the terms in a many to many relationship).  The concept definition (in the Notes field) is separated into two parts: a normative (the proverbial 25 word or less) definition and a more discursive and explanatory discussion.  We only move the normative definition into the glossary.  Where a term has multiple meanings, we aggregate them fro the individual concepts.
We do the Glossary population via MS Access macros and some VB code (to do the normative definition extraction and merging.)

It wasn't that hard, once we'd sorted out what needed to be done.

We maintain the different forms semi-automatically.  When we create new terms a separate macro is run to add indicative values for the different forms using some simple rules and then we check the values to make sure they're OK (joined against the terms added since last inspection).

Does that help?

Triplicating the definitions is ugly, but needs must.

Paolo

[Eve]

This actually sounds very interesting. Not because I have a customer asking for it, but simply out of personal interest.

Is this proprietary or something you could communityize? communify? communalize? release?

There are publicly available ontologies.

One is Suggested Upper Merged Ontology (SUMO). In addition to being downloadable you can query it with respect to English words from:
http://54.183.42.206:8080/sigma/Browse.jsp?kb=SUMO

[Glassboy]

, the cost/benefit ratio is simply far too high.

No. As I said the current highlighting is not done via a semantic analysis but by simple string comparison. So just add the "s follows string" rule to underline also the "s". Simple as that. Works in more than 90% of all cases I'd say at almost no cost.

q.

Your scheme fails the moment someone forgets an apostrophe on a possessive or uses something that approaches being a homograph, or even uses the -ies -ves forms of the plural.

EDIT:  And you'd catch the past participle form of verbs.

[AndyJ]

and while we're talking about the vagaries of the English  language, a topic that is dear to my heart...

Pea (the vegetable) is a back formation from Pease (the plant) which now has a plural peas (the vegetable), and the plant itself is now also called pea. The back-formation occurred because people mistakenly thought that pease was the plural for the little green balls...

But, back on topic, an extra field, which allowed synonyms for a term to be recorded by the user would really help!

[Paolo F Cantoni]

and while we're talking about the vagaries of the English  language, a topic that is dear to my heart...

Pea (the vegetable) is a back formation from Pease (the plant) which now has a plural peas (the vegetable), and the plant itself is now also called pea. The back-formation occurred because people mistakenly thought that pease was the plural for the little green balls...

But, back on topic, an extra field, which allowed synonyms for a term to be recorded by the user would really help!

There is no concept of "On  premise" - the word is not valid (technically, I believe it's a non-sequitur) and is incorrectly used by people who don't actually understand the language they are using.

We don't enter the premise, or abscond from the premise.  When referring to buildings the term is "the premises" - thus: on premises.  If you don't know why, then look it up.

What happened was a whole pile of people who didn't understand the language couldn't/wouldn't correct each other.

We can't build systems that work because we can't tell each other unambiguously what we want.

My rant for the day...

Please try NOT to use words whose meaning you don't know.

Paolo

[AndyJ]

While we are griping...

We have vehicles, covered with livery (lights, sirens, checkered stripes etc).

They are referred to as "fleets".

As in, grab a couple of "fleets" from the pool for Saturday morning...

Not kidding, one fleet, two fleets. (Drives me crazy.)

If that's not bad enough, this organisation has preserved a somewhat archaic meaning of "duplicitous"...

Around here, duplicitous means doing something twice, with no intention of implying that something is underhanded.

But, I would really, really like an extra column in the glossary, where the user gets to provide an alias for a term.

Andy