[tlug] Re: [A] Using [TAGGING] in email subjects

["emiddleton@example.com" <emiddleton@example.com>]

Keith Bawden wrote:
> On 8/7/07, emiddleton@example.com <emiddleton@example.com> wrote:
>   
>> Keith Bawden wrote:
>>     
>>> simply encouraging people to use descriptive subject lines should
>>> suffice.
>>>       
>> writing descriptive lines does little for searching.  Just because you
>> have a particular word in the subject doesn't mean the subject concerns
>> that word.
>>     
>
> No but a nicely written subject will help more than a cryptic tag. For example:
>
> [tlug][IME] Done gone been broke, and my foot hurts.
>
> Or
>
> [tlug] SCIM is not working on my Debian Etch box
>   

but

[SCIM] is not working on my Debian Etch box

would be even better because it would distinguish the topic as being about SCIM as apposed to

could someone recommend an alternative to SCIM


>>> After all I know I wont be posting much if I have to do a
>>> table lookup in order to craft a well formed, fully tagged subject
>>> line. What a PITA.
>>>       

>> If you have to do a table lookup then the tags weren't selected very
>> well.  Which is why I would prefer to flesh out the idea before
>> committing to a base set.
>>     
>
> If we do not consistently use what ever tags we decide upon then their
> effectiveness is limited.

Yes, if we used [SCIM] to refer to my app named scim as apposed to the
input method then this would make the tags less effective.

>  If we select tags based on "common sense"
> then someone will need to go out and research what is common sense and
> who actually has it - if it is found to exist.
>   

Common usage not common sense.

> Another way of puting it. If we have to think hard, debate, and
> "flesh" out this list of tags and the associated descriptions of the
> type of information the represent then it would suggest such a list
> would not be immediately useful without everyone doing table lookups
> when crafting the subject line to an email.
>   

In 'fleshing out' a forksonomy we have found that IME, FEP, HENKAN are
not naturally associated with input methods by a number of people, but
the direct application reference SCIM is.

Edward