On Friday 04 November 2005 8:11 pm, [EMAIL PROTECTED] wrote: > Dear David, > > I know Wordnet and I have played with it for a while. I also had a > look at other "professional" bibliographic thesaurus, such as the > CurrentContents Thesaurus. > The problem is that people working in different subject areas have > different points of view regarding the "correct" hierarchy. > > I agree that it would be very elegant to be able to use something > like Wordnet as Thesaurus. > However personally I would prefer to be able to create "my" thesaurus > myself. I think that perhaps the choice of the word "thesaurus" was > not a very good one. > I think rather that simply the keyword database should be relational, > in the way that each keyword has a number and in the database are > stored with each keyword the numbers of the keywords that are > synonymnous (e.g. 0) and the numbers of the keywords that are one up > in the hierarchy (e.g. 1) (subjects / topics ?) (down in the > hierarchy is not interesing for searching). > > 1 plant > 2 tree,1,1 > 3 beech,4,0,2,1 > 4 Fagus,3,0,2,1 > 5 Fagus sylvatica,3,0,4,0 > etc ... >
> The rudimentary user interface could be simply 2 drop down lists side > by side with the keywords and between the two lists two checkboxes, > one that says "..is also called..." for synonyms and one that says > "... is a ...." for subjects, and a save button. And perhaps one > "output" button that shows the list including for each keyword the > synonyms and topics. > By the way all of this reminds me a little bit of an old sourceforge > PalmPilot project that was called ThoughtStream : basically here > notes were organised by associations and links. > Or also the theory of Semantic networks.I think that there are > several projects out there that have already produced code for this. > > I'm not really a programmer, I mainly do some macro programming, > however if the bibliographic project is really interested in having > linked/hierarchical keywords I could try to think more about how to > realize the input/output/searching in theoretical and UI terms. > I am interested and I think it would be a good idea. If you work on a description I am willing to set up a web page for it - to help stimulate discussion and to make sure we do not forget it. > Should I switch to the developer list ?? > Yes, it is much more active. > Cheers > > Oliver > > On 4 Nov 2005 at 7:53, David Wilson wrote: > > Oliver, > > > > Your suggestion sound very usefull. > > > > Have you considered how thesaurus software might be used to provide > > support for this. Wordnet is a thesaurus database which (I think) has > > been used as the source for the OOo English thesaurus. > > > > The OOo English thesaurus gives for "beech" gives "Beech Tree" and "Tree" > > > > A wordnet query gives as one of its several query type results - > > > > > > > > Sense 1 > > beech, beech tree -- (any of several large deciduous trees with rounded > > spreading crowns and smooth gray bark and small sweet edible triangular > > nuts enclosed in burs; north temperate regions) > > MEMBER OF: Fagus, genus Fagus -- (beeches) > > > > So it would seem that wordnet even provides a latin name list. > > > > Worknet also contains a hierachy of terms - > > > > beech, beech tree -- (any of several large deciduous trees with rounded > > spreading crowns and smooth gray bark and small sweet edible triangular > > nuts enclosed in burs; north temperate regions) > > => tree -- (a tall perennial woody plant having a main trunk and > > branches forming a distinct elevated crown; includes both gymnosperms and > > angiosperms) > > => woody plant, ligneous plant -- (a plant having hard > > lignified tissues or woody parts especially stems) > > => vascular plant, tracheophyte -- (green plant having a > > vascular system: ferns, gymnosperms, angiosperms) > > => plant, flora, plant life -- (a living organism > > lacking the power of locomotion) > > => organism, being -- (a living thing that has (or > > can develop) the ability to act or function independently) > > => living thing, animate thing -- (a living > > (or once living) entity) > > => object, physical object -- (a tangible > > and visible entity; an entity that can cast a shadow; "it was full of > > rackets, balls and other objects") > > => entity -- (that which is perceived > > or known or inferred to have its own distinct existence (living or > > nonliving)) > > > > Regards > > > > David > > > > On Friday 04 November 2005 12:20 am, [EMAIL PROTECTED] wrote: > > > Hello Biblio-Developers, > > > > > > I'm a researcher in natural sciences and I have been using > > > bibliographis software for a several years now, startig with the > > > first version of Reference Manager, recently switching to EndNote > > > (with some regrets), but I have also been testing several > > > share/freeware programs. > > > My expectations are rather humble, so most programs would do most > > > things I like them to, why I got stuck with EndNote is rather a > > > problem of easy compatibility with colleagues databases. > > > I have read the basic papers on the web page of the bibliographis > > > project and I have not much to add (well, there are always details on > > > a which list .... :-), however there is one thing I would find > > > immensely useful and have to date not found in any program I tested. > > > And this is hierarchical keywords > > > Let me explain with some example from my subject : > > > - you might have several references in your list that treat > > > different tree species : oak, beech, pine, spruce. > > > now two of these are deciduous (oak , beech) and two are coniferes > > > (pine, spruce), and all of them are forest trees. > > > What I would like a bibliographic program able to do is to look up > > > the keyword "beech" in a thesaurus (that the user created personally > > > for his keywords) and find that its a deciduous and a tree. With the > > > effect that I will add to the references the keywords for beech, oak > > > etc. and when I search for "deciduous" I will find all articles > > > concerning beech and oak and when I search for "tree" I will find all > > > references with keywords "oak", "beech", "pine", "spruce". > > > Basically this consists in linking keywords unidirectionally with one > > > another. > > > However also linking bidirectionally is (or even more) important : > > > same example : in one reference concerning oaks I type "oak" as > > > reference, in another i might type "Quercus", which is the latin name > > > for oaks; When I want to search for references concerning oaks, I > > > always need to remember that there were two possible ways to code the > > > keyword for this reference and search both. With this simple example > > > this might seem not a difficult problem, in a conventional program I > > > just need to do a search of "oak" OR "quercus". However the problem > > > is that in a scientific reference list that is used for bibliographic > > > work (that is not only "citing" but mainly searching), there will be > > > many keywords as example one of my main reference data bases has > > > about 2000 references abd nearly 3000 keywords). Many of these > > > keywords will be synonymous and its easy not to remember one of the > > > synonyms in a search and thus not finding certain articles. > > > Therefore I would appreciate much if the keyword database of the > > > bibliographic project could have the ability to be relational so that > > > one could easily link uni o bi directionally several keywords. > > > > > > I would be happy to discuss this more with anybody who has any ideas > > > on this subject. > > > > > > Cheerio > > > > > > Oliver Brendel > > > > > > > > > _________________________________________________________ > > > Oliver Brendel UMR Ecologie et Ecophysiologie Forestiere > > > INRA - Centre de Nancy 54280 Champenoux France > > > [EMAIL PROTECTED] TEL 0033/383/394041 ext4201 FAX /394069 > > > > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > > For additional commands, e-mail: > > > [EMAIL PROTECTED] > > > > -- > > ------------------- > > David N. Wilson > > Co-Project Lead for the Bibliographic > > OpenOffice Project > > http://bibliographic.openoffice.org > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] -- ------------------- David N. Wilson Co-Project Lead for the Bibliographic OpenOffice Project http://bibliographic.openoffice.org --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
