[Ontology-editors] use of word "gloss" Re: plant-type terms etc.

Chris Mungall cjm at berkeleybop.org
Tue Jun 2 15:35:01 PDT 2009


On Jun 2, 2009, at 10:00 PM, Karen Christie wrote:

> Interesting that there is actually a definition that is non- 
> pejoritive and relevant to the context it is being used in here,  
> though from Amelia's comment it seems that we are using too much  
> text for that meaning to actually be appropriate.
>
> I find the word offensive because when I look it up on my default  
> web based dictionary (www.dictionary.com), these are the defs at the  
> top of the page, several of which use words such as 'false' or  
> 'superficial', though there is also one def relating to the meaning  
> Amelia isn't so fond of ;)
>
> noun
> 1. 	a superficial luster or shine; glaze: the gloss of satin.
> 2. 	a false or deceptively good appearance.
> 3. 	Also, glosser. a cosmetic that adds sheen or luster, esp. one  
> for the lips.
> verb (used with object)
> 4. 	to put a gloss upon.
> 5. 	to give a false or deceptively good appearance to: to gloss over  
> flaws in the woodwork.
>
> And responding to the question, do I find the 'logical' definition  
> to also be descriptive, no, I don't. I find the logical def to be a  
> text statement of things I can see from the graph, but it does not  
> describe the relevant biology to me.

Actually there are very few logical definitions that are replicated in  
the graph at the moment, unless you are looking at the full GO + xp  
set, and even then that's just for the more trivial compositional  
terms (sorry, "trivial" is another of those words that has a technical  
CS sense and a potentially pejorative one). At the moment every  
relationship in GO is a necessary conditions. Definitions, by  
definition, are necessary and sufficient conditions.

> I'm not tied to the word 'descriptive' if someone has another  
> suggestion.

How about this. The definition tag contains descriptive text which  
consists of the following components (all optional except the first,  
and the second strongly recommended)

- a definition - necessary and sufficient conditions, ideally  
specified in genus-differentia form, where the genus is an is_a parent  
or ancestor and the differentia are relationships to other GO terms or  
other OBO ontologies
- explanatory text - expanded form of the definition containing  
additional descriptive text or explanations clarifying the definition
- examples
- counter-examples
- additional notes

>
> thanks,
>
> -Karen
>
>
> On Tue, 2 Jun 2009, Chris Mungall wrote:
>
>>
>>
>> On Jun 2, 2009, at 6:52 PM, Karen Christie wrote:
>>
>>> Could we please stop using the word "gloss". It is completely  
>>> meaningless.
>>
>> Merriam Webster:
>> 1 a (1) : a brief explanation or a translation or definition (as  
>> one appearing in the margin or between the lines of a text or in a  
>> wordbook based on the text) of a textual word or expression felt to  
>> be difficult or obscure (2) : an expanded interpretation of or  
>> commentary on a textual word or expression
>>
>>> For the "logical" part of the definition, we have a word that  
>>> explains the type of information. "Gloss" does not do that. Could  
>>> we start calling this the "descriptive" part of the definition.  
>>> Then both aspects of the definition have a word that is  
>>> descriptive as to the type of information contained there.
>>
>> sure, but would you not consider the logical part also descriptive?  
>> I'm not sure why you find the word gloss so offensive. I'll try and  
>> stick with "descriptive part of definition" but I am a very slow  
>> and lazy typist so sorry in advance if I occasionally slip into  
>> saying "gloss"
>>
>>> -Karen
>>> On Tue, 2 Jun 2009, Jennifer Deegan (nee Clark) wrote:
>>>> Great, thanks. If I don't hear any objections I'll start on  
>>>> Monday then.
>>>> Jen
>>>> Jane Lomax wrote:
>>>>> Yep - I'd just go ahead. I'll be very glad to see the back of  
>>>>> sensu!
>>>>> Jane
>>>>> Jennifer Deegan (nee Clark) wrote:
>>>>>> Excellent, thanks. If I hear no other objections I will plan to  
>>>>>> start this coming Monday.
>>>>>> Are you happy for me to do this without consulting people  
>>>>>> outside this list?
>>>>>> Jen
>>>>>> Midori Harris wrote:
>>>>>>> no objection, then
>>>>>>> m
>>>>>>> On Mon, 1 Jun 2009, David Hill wrote:
>>>>>>>> I think if Jen adds the appropriate examples to the  
>>>>>>>> definition gloss, we could live without these.
>>>>>>>> Jennifer Deegan (nee Clark) wrote:
>>>>>>>>> That might be nice actually, but I'd still like to make sure  
>>>>>>>>> that the terms stand alone without these synonyms. I think  
>>>>>>>>> it is very easy for those of us within the GO to rely on  
>>>>>>>>> these synonyms as clues, and the other users  have to manage  
>>>>>>>>> without. I think if we are not prepared to live without them  
>>>>>>>>> then we need to think carefully about why that is.
>>>>>>>>> Jen
>>>>>>>>> Chris Mungall wrote:
>>>>>>>>>> What about adding a synonymtypedef 'historic', or  
>>>>>>>>>> 'deprecated'?
>>>>>>>>>> On Jun 1, 2009, at 9:07 AM, Jennifer Deegan (nee Clark)  
>>>>>>>>>> wrote:
>>>>>>>>>>> Hi,
>>>>>>>>>>> I think that for new users it is confusing to have them in  
>>>>>>>>>>> place. We no longer explain to users what 'sensu' means  
>>>>>>>>>>> and they are likely to look at the synonyms and assume  
>>>>>>>>>>> that they are still current and ascribe their own assumed  
>>>>>>>>>>> meaning to them. I think it is very important that they  
>>>>>>>>>>> are either still supported and kept correct or removed  
>>>>>>>>>>> altogether.
>>>>>>>>>>> Jen
>>>>>>>>>>> David Hill wrote:
>>>>>>>>>>>> Jen,
>>>>>>>>>>>> How does it hurt to have the synonyms there?
>>>>>>>>>>>> David
>>>>>>>>>>>> Jennifer Deegan (nee Clark) wrote:
>>>>>>>>>>>>> Hi,
>>>>>>>>>>>>> Following on from this, I would like to start stripping  
>>>>>>>>>>>>> out the sensu synonyms from the live file.
>>>>>>>>>>>>> I am not planning to do it in one big deletion, but  
>>>>>>>>>>>>> rather remove them incrementally, checking in each case  
>>>>>>>>>>>>> that no information is lost. I envisage going through  
>>>>>>>>>>>>> the terms on at a time, adding examples as discussed  
>>>>>>>>>>>>> here, and deleting the synonym when I am happy that the  
>>>>>>>>>>>>> meaning of the term is clear. I do not anticipate that  
>>>>>>>>>>>>> this will take very long.
>>>>>>>>>>>>> Would anybody have any objection to my starting to do  
>>>>>>>>>>>>> that?
>>>>>>>>>>>>> Thanks,
>>>>>>>>>>>>> Jen
>>>>>>>>>>>>> Midori Harris wrote:
>>>>>>>>>>>>>> Yes, taking the issue of examples separately, there's  
>>>>>>>>>>>>>> no reason not to include more, and they would probaby  
>>>>>>>>>>>>>> be very helpful.
>>>>>>>>>>>>>> m
>>>>>>>>>>>>>> On Mon, 1 Jun 2009, Chris Mungall wrote:
>>>>>>>>>>>>>>> I agree, no need to revisit the naming issue.
>>>>>>>>>>>>>>> However, these terms are in the minority. There are a  
>>>>>>>>>>>>>>> larger set of terms in which specific examples would  
>>>>>>>>>>>>>>> be helpful. Examples would always go in the gloss part  
>>>>>>>>>>>>>>> of the definition, and ideally have a citation. I  
>>>>>>>>>>>>>>> think this is just clarifying existing GO policy.
>>>>>>>>>>>>>>> On Jun 1, 2009, at 8:47 AM, Midori Harris wrote:
>>>>>>>>>>>>>>>> I seem to recall several annotators expressing a  
>>>>>>>>>>>>>>>> strong preference for the names like 'plant-type  
>>>>>>>>>>>>>>>> vacuole' when we discussed it at a GOC meeting. Given  
>>>>>>>>>>>>>>>> that these names got the full discuss-at-meeting  
>>>>>>>>>>>>>>>> treatment, and that the current names are thus a  
>>>>>>>>>>>>>>>> consensus meeting outcome (tm), I wouldn't change  
>>>>>>>>>>>>>>>> them without obtaining explicit approval from the  
>>>>>>>>>>>>>>>> larger GO group.
>>>>>>>>>>>>>>>> m
>>>>>>>>>>>>>>>> On Mon, 1 Jun 2009, Jennifer Deegan (nee Clark) wrote:
>>>>>>>>>>>>>>>>> Hi,
>>>>>>>>>>>>>>>>> A proposal has been made as follows:
>>>>>>>>>>>>>>>>> Instead of putting modifiers like 'plant-type' in  
>>>>>>>>>>>>>>>>> term names to make the meaning of the more  
>>>>>>>>>>>>>>>>> esoterically defined terms clear, we should just put  
>>>>>>>>>>>>>>>>> an example in the definition gloss. For example this  
>>>>>>>>>>>>>>>>> term:
>>>>>>>>>>>>>>>>> GO:0000325
>>>>>>>>>>>>>>>>> name: plant-type vacuole
>>>>>>>>>>>>>>>>> exact: vacuole, cell cycle-independent morphology
>>>>>>>>>>>>>>>>> def: A closed structure, found only in eukaryotic  
>>>>>>>>>>>>>>>>> cells, that is completely surrounded by a unit  
>>>>>>>>>>>>>>>>> membrane, contains liquid, and retains the same  
>>>>>>>>>>>>>>>>> shape regardless of cell cycle phase. [source:  
>>>>>>>>>>>>>>>>> GOC:mtg_sensu, ISBN:0815316208]
>>>>>>>>>>>>>>>>> would become:
>>>>>>>>>>>>>>>>> GO:0000325
>>>>>>>>>>>>>>>>> name: vacuole, cell cycle-independent morphology
>>>>>>>>>>>>>>>>> def: A closed structure, that is completely  
>>>>>>>>>>>>>>>>> surrounded by a unit membrane, contains liquid, and  
>>>>>>>>>>>>>>>>> retains the same shape regardless of cell cycle  
>>>>>>>>>>>>>>>>> phase. An example of this structure is the vacuole  
>>>>>>>>>>>>>>>>> of plant cells. [source: GOC:mtg_sensu, ISBN: 
>>>>>>>>>>>>>>>>> 0815316208]
>>>>>>>>>>>>>>>>> Would anybody have any objection to this change? The  
>>>>>>>>>>>>>>>>> advantages are that this policy would remove  
>>>>>>>>>>>>>>>>> prominent taxon information from the file but would  
>>>>>>>>>>>>>>>>> still leave information in the gloss to clarify the  
>>>>>>>>>>>>>>>>> meaning of the term. The disadvantage is that the  
>>>>>>>>>>>>>>>>> less prominent clue to the meaning of the term would  
>>>>>>>>>>>>>>>>> leave users having to hunt a bit more to find what  
>>>>>>>>>>>>>>>>> they want.
>>>>>>>>>>>>>>>>> Thanks,
>>>>>>>>>>>>>>>>> Jen
>>>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>>>> Ontology-editors mailing list
>>>>>>>>>>>>>>>>> Ontology-editors at geneontology.org
>>>>>>>>>>>>>>>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>>>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>>>>> Ontology-editors mailing list
>>>>>>>>>>>>>>>> Ontology-editors at geneontology.org
>>>>>>>>>>>>>>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>>>>>>>>>>>> _______________________________________________
>>>>>>>>>>>>> Ontology-editors mailing list
>>>>>>>>>>>>> Ontology-editors at geneontology.org
>>>>>>>>>>>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>>>>> _______________________________________________
>>>>>> Ontology-editors mailing list
>>>>>> Ontology-editors at geneontology.org
>>>>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>>> _______________________________________________
>>>> Ontology-editors mailing list
>>>> Ontology-editors at geneontology.org
>>>> http://fafner.stanford.edu/mailman/listinfo/ontology-editors
>>
>



More information about the Ontology-editors mailing list