search

climsoft / Climsoft

Climsoft Desktop for Windows - http://www.climsoft.org
GNU General Public License v3.0
14 stars 28 forks source link

Overriding of data in the database in data entry and metadata forms #439

Open maxwellfundi opened 5 years ago

maxwellfundi commented 5 years ago

We will be putting things in place for the data entry forms not override data in the tables because this will be part of the QC, by keeping a record of all previous values. Should the metadata forms also do this storing of previously entered values?

smachua commented 5 years ago

@maxwellfundi and @mhabimana sorry I hadn't much time to look at the activities happening on the metadata forms. I highly welcome any necessary improvement on metadata management but I wish we can have it well planned and structured.

It's important to note that the description of the observations in the database is highly dependent on the underlying metadata. Hence any change on how the metadata are being managed mainly through the current forms and other dialogs should be well thought of. To be sure that it is done right I propose that we hold a Skype meeting and agree on the following:

(i) The areas of metadata management that require improvements and how it should be done, (ii) The benefits to be gained, (iii) The impacts on the current operations, (iv) Who is to do what? (in the spirit of team work).

If my proposal is agreeable then I suggest that @mhabimana organizes the meeting that at least includes 3 of us.

Steve-Palmer commented 5 years ago

History of metadata changes is important. Note that there is a question circulating in WIGOS about how to handle linking stations which could be chained together to make a longer record, such as where a station has moved. For example, Entebbe Botanical Gardens has a record from 1901. Later Entebbe Airport opened. To my knowledge, the airport observing station has had at least four locations within the airport area. And then you get changes in instrument systems and locations even when the base station has not moved. All of these need to be recorded with start and end dates.

maxwellfundi commented 5 years ago

@smachua Thank you for your comment on this issue.This issue is just a thought on a possible future improvement and we would first have to broadly consult before such a change is done. Currently, we don't see it as an urgent task so a discussion could wait until there is interest to implement this feature.

shadrackkibet commented 5 years ago

Good to see that this issue has been raised here and at least discussed. I am currently working on my project which involves homogenization of climate records here a KMD. I find metadata information that is available not very reliable especially when it comes to when the station opened or closed. For instance, for the year when the station closed/opened which sometimes do not correspond with the data that is available. I also find that giving the year when the station opened /closed is not sufficient, it is limited. I would like to know exactly when the station started (Jan. Feb, Mar.......). Keeping a record to show when a station moved to be considered. Another thing I find problematic here is station names, some stations have several names(aliases) which should be part of metadata as well. I am aware WMO has a guide on climate metadata and homogenization. I think generally climsoft should meet the minimum requirement WMO recommends to store and easily manage this kind of metadata.