User:Anwar saadat/bubble maps (FAQ)

Why do we need a bubble map?
Traditional colour-coded maps are useful to represent relative data (like per capita GDP). But, they are unsuitable to represent absolute data (like GDP) as they depend on arbitrary cut-off ranges (like shading all economies in the range $100 billion to $1 trillion as blue!). Unless a continuous and evenly-changing colour scale is used. Another problem with colour-coded maps is that countries with large area but low population (Greenland, Antarctica) get undue prominence for economic measures.

When is a bubble map unsuitable?
Bubble maps have weaknesses too. They do not represent the distribution of output within a nation. Readers ought to understand that these are not replacements for precision satellite imagery. Also, the top producer is always shown with a singular large green bubble. Its significance as "locus of control" (or "poster boy" or "eye of the storm" or "symbol of stability") within the industry must not be missed. Bubble maps are unsuitable to represent relative data (like per capita GDP). In JPG and PNG versions, yellow bubbles (in highly competitive industries) tend to spill out of the national borders onto the ocean or neighbouring territories. This is rectifiable in SVG versions. Bubble maps do not reveal insignificant producers in a industry (whose output is less than 1 per cent of the topper). So many poor African nations may never be represented. However, this is rectifiable by introducing a new blue-bubble set weighted at 0.1 per cent of the top producer.

Is a bubble map same as a density map?
No. On the surface, the reader may not see much of a difference. But there are significant conceptual differences. Unlike a density map, a bubble map ties together all data in a set through ratios pegged to the top producer making it consistent even in incomplete-data-set environments. It means, you know instantly who is the top dog in the respective industry as well as the "balance of power" between any two producers. This is closer to how the real world operates where each manufacturer (including a new start-up) is immediately aware of his largest competitor but almost always clueless about the actual number of his competitors. All business decisions are constrained by this uncertainty in any industry. Readers must therefore keep in mind that these bubble maps represent the spread not actual volume or valuation.

Which colour and size is suitable for the bubbles?
Colour theory stresses on both usability as well as aesthetic appeal. Red, Green and Blue are the staples in any modern colour scheme. The remaining 16.5 million colours are arithmetic derivatives of these three colours. This editor adopted the traffic-sign model and inserted a intermediary Yellow (instead of Blue). These three colours can be rendered even in pretty old dusty legacy computer screens.

The green bubble (32 x 32px) represents the top producer in the industry. The yellow bubble (16 x 16px) represents the major producers (weighted atleast 10 per cent of the green bubble). The red bubble (8 x 8px) represents the minor producers (weighted atleast 1 per cent of the green bubble). Size differentiation resolves the accessibility issues faced in old computer screens. Else, the reader may see only shades of Black and White.

It should be noted that in several industries like zirconium or ilmenite, the reader may see multiple green bubbles. No cause for alarm. It simply means the respective industry is two-headed or three-headed as per case. The nations with green bubbles are, broadly speaking, price setters and dictate terms to the rest of the industry.

Which file extension is suitable for such bubble maps?
Typically a JPG version would cost 500KB and a PNG version 50KB in disk space. But this editor is planning to move all bubble maps, without any courtesy of warning, to SVG version by the end of 2008.

It should be noted that the SVG format was initially targeted at the print media market only in circa 2000. But by 2005, it became popular even in Web, handheld device and other screen markets (challenging the historical dominance of JPG and PNG). A typical SVG version of a bubble map offers layers, infinite scalability with average file size reduced to 20KB provided its author faithfully optimised the file by object re-use (like circle, line, label,...) and validated XML tags for formation with third-parties like W3C. SVG maps may take longer to create compared to JPG and PNG versions. But they are easier to maintain. For instance, suppose Tibet gets independence tomorrow, then the political map of the world need to be changed. It means, your entire work has become obsolete overnight if you use JPG or PNG output maps. Instead, if you use a SVG version, replace the blank world map in the respective layer only and re-upload within minutes. Also, bubbles can be individually moved and manipulated after upload by any other editor (apart from original author) after receiving new data. It means updates are quicker than with static JPG or PNG maps. No need to re-work the whole map again form scratch.

But there is a catch with SVG too. Microsoft, the top dog in client PC market, does not support SVG by default (unlike JPG and PNG). So, the 684 million annual visitors to Wikipedia may stand a real risk of staring at white space in each image namespace.

Which software is suitable to create a bubble map?
JPG and PNG versions can be easily created independently with utter-gutter software like Microsoft Paint. SVG needs freeware like Inkscape.

How to create a bubble map independently?
First, you select the data from the source (like GDP statistics from imf.org) for download. You would need a spreadsheet software (like Microsoft Excel) to do the formatting, sorting, totalling, percentages and rounding off. Use any of the above software for mapping. Before uploading to Wikipedia, please ensure that you have optimised the filesize by chopping off redundant pixels. As a rule of thumb, stick to above filesize limits.

Can you upload the bubble maps to the Commons?
No. Though it is tempting to enter history, it is advised that readers too refrain from adding the bubble maps seen on article namespace to the Commons. This is because, these are not blank templates. Also, these maps are updated annually or adhoc whenever new data is discovered, whichever earlier.