[Live-demo] Re: [OSGeo-Discuss] Re: [Gvsig_english] New site demonstrating the use of Free & Open Source Software

[Simon Cropper]

Helena,

Thanks for the comments. I have included some feedback regarding 
particular points below.

On 30/04/11 11:53, Helena Mitasova wrote:
> I have prepared this data set because I thought that it would be useful
>   to have a set of simple data sets for different regions with standardized names of data layers
> so that we can use the same tutorial with data sets from different regions.

I understand the concept but if you did this you would need to 
standardise the field names also.

> So Simon, if you are going to create a data set for Australia to go with your tutorials,
> it would be great if you could use the same names as in our basic data set (or if you have a suggestion
> for a different name, please let me know - at this stage it will be easy to change it on
> our side).

I have already acquired some local data. You can see what is provided at 
http://gis.fossworkflowguides.com/#data.

> Then we can use your tutorial with our nc data set and students/users
> in Australia can use our tutorials with your data set. We can have this for many countries
> and many different software packages we just need to agree on the names for data layers.

I can see the logic here. Standard file names. Standard attribute names. 
Various formats. Various regional datasets.

How do you propose to store different languages? This would 
intrinsically change both the file and field names.

When I considered localisation and translation of my tutorials, I though 
that most people would just recreate the images and rejig the text 
accordingly - using the existing file as a predominantly completed template.

> There are certain tasks that are region specific, such as the coordinate systems,
> but many tasks, from display to analysis, would be the same.

With this in mind, putting aside regional aspects, having one dataset 
that most people use in things like the LiveDVD provides for a 
consistent experience. My only issue here is the use of low resolution 
continent-wide data in the quickstarts and tutorials, when most people 
would be working at a regional or local level.

Regional aspects is an interesting issue that need further 
consideration. Apart from distinct CRS, "local" files have distinct 
names, fields of attribute tables have distinct names, data has varying 
accuracy and/or currency, etcetera. These variations make working with 
regional datasets unique and tutorials that demonstrate the use of 'raw' 
files -- as you would expect to get them if you went and purchased them 
from a regional authority -- valuable and enlightening.

With this in mind I must confess I am in two minds with trying to 
standardise all data. When considering creating a tutorial series I 
wanted to provide a resource for all people to use regardless of the 
country. In my mind this only required me to provide the data used to 
create the tutorials so others could repeat the steps -- which I did -- 
and ensure that colloquial terms are adequately explained.

Derivative creation is another issue intrinsically bound to this issue. 
Apart from slight variation in text (assuming you are not translating 
the document) all the images need to be redone to show the local data in 
context. If you are going to go through all this effort changing a few 
names is a minor issue. The way I have handled this in my tutorials is 
to tag file names and other elements with unique tags. This provides the 
ability to substitute elements of the HTML webpage based on a simple 
translation table and substitution routine. In theory, it would be a 
simple matter stating that when using this vector file the term altitude 
should be replaced with elevation.

My experience has show that the biggest impediment to derivative 
creation is the ability to disarticulate a tutorial, modify those 
elements that need changing and put it back together quickly -- as 
explained in detail on my website, this is why I fell back to HTML.

Ideally, we should have a Content Management System that stored 
educational material at high enough resolution to manage this 
disarticulation quickly and efficiently; but, alas one does not exist at 
present and it is necessary to hobble together an alternative.

So in summary, I can't see the value of modifying the current data I 
have to make a consistent dataset, as creation of a derivative using the 
North Carolina dataset will require recreation of the 30-40 images shown 
in a tutorial anyway, and a simple search-and-replace of a dataset name 
or field name would be a minor additional task.

-- 
Cheers Simon

    Simon Cropper
    Principal Consultant
    Botanicus Australia Pty Ltd
    PO Box 160, Sunshine, VIC
    W: www.botanicusaustralia.com.au