12 January 2007
Uncertainty:
"We cannot always be certain about things. This is as true of geography as it is other aspects of life." - we might consider adding to this a pithy quote about the uncertainties of life from some ancient source (Thucydides has stuff on the uncertainties of war, but surely there are others).
"In developing our data models and interfaces for Pleiades -- and in conducting editing of data by the community -- we must take stock of the varied ways in which uncertainty affects us. We must avoid giving the false appearance of exactitude while maintaining ease of use." - Very important that we make this clear. It is part of the Pleiades system that we strive for scholarly accuracy - beyond multiple user accuracy of other Wiki-sites. I wonder if we might change "data models and interfaces" to something a bit more simple: ie. "maps and indexes" or similar. It seems to me to come down to a question of how sophisticated a language we want to use and who our audience will be.
The section on Names is good and clarifies a number of my lingering questions. "Names ΒΆ
- Accuracy of attestation
- Accurate: string of characters in witness is thought to be correct (BAtlas: no special indication)
- Inaccurate: witness is thought to be incorrect (BAtlas: name enclosed in inverted commas)"
- I assume what is meant here is that the names we will read will be "considered" as "accurate" or "inaccurate", etc. I might consider a short sentence introducing this section saying that; e.g. "The certainty of names is divided into three categories (outlined below) ..." or some such thing. I think it might just be a question of transition. - also regarding the names section: the descriptions and the chart just barely fit on my rather large computer screen. Users with smaller screens will be forced to scroll up and down to compare example with explanation. Why not just included the example with the explanation. This may not be possible with all the information we want to get across, but I think we need to find a way to help the user in this case.
"string of characters in witness is thought to be correct" - by witness do we mean "source" - witness is confusing, especially if our only evidence comes from an inscription.
Locations:
"Positional accuracy (horizontal)
- Definition: the degree to which information on a map or in a digital database matches true or accepted values (see Geographer's Craft, s.v. "Error, Accuracy and Precision")
- See separate discussion: DataPrecisionAndAccuracy "
- I appreciate the link here, and it is very helpful. Still, we might consider lengthening our definition and explanation so that only the most interested user will have to read the linked article. For example, we may want to explain what is meant by horizontal immediately.
- the definition for "positional precision" is far to vague (for me anyway)
- Regarding the Placement certainty - is there a reason for our changing the format of the Batlas? Note also that the inclusion of the map makes this section extremely clear and we may consider including such an example in the Names section. Actual examples, i.e. "screen captures" (I think that is what they are called), of the Pleiades maps with names included, might help explain the system along with the chart.
"Associations of Names and Locations"
"Pleiades models the relationship between location and name slightly differently, as follows." - sorry to be a nuisance, but this sentence, especially the "slightly differently", just sounds odd.
- I understand what is going on in the rest of this section but I think we might clarify it a bit more. For example I have no clue to what "1..n" refers.
On the whole this is all very exciting. I am excited to see things coming together and taking shape.
