My reflections:
1. Indeed, every bit of data available on the net is 'good' or 'bad' depending upon who uses it and in what context and perspective. Security concerns about infrastructure should always be a cause of worry in our times and circumstances.
However, given the conspicuous nature of geodetic presentations available today, (including google/yahoo/ MSN earth maps and BHUVAN (India's ISRO project)), for someone with enough serious bad intentions, it is not at all difficult to gather and accumulate such data. Whether it is power lines or water distribution systems or even an army corps HQ, it is all out there open luckily or unluckily! Perhaps, more than any network structures (like a railway line or road system or canal system), the most obvious man-made features on an earth view at 1 to 5 kilometer zoom level, are High Tension lines!
2. I presume that many of the features that we add now into google mapmaker, will ultimately end up within layers of different functionality. Although they are all seen together at present in the same mapmaker plain, it will be easy for Google to segregate them into different thematic layers or vectors at any point of time.
Depending upon the nature of mapmaker features, their application and usefulness can be varied. For eg: the electrical network of a national power grid is useful mostly only for power system engineers and country/city planners. But for them, indeed, these are immensely useful data. I personally know situations where, (before a decade or so) had such data been available to the less privileged members of design teams, some huge historic blunders would not have taken place!
3. In fact, in real life situations at remote locations, some of the easiest reference feature to cross check with, are these very structures. Whether it is a desert in Middle East or a cliff on the rain forests of Kerala, with just a GPS unit and Google (-like) map, if you come across a power line, I bet that will be the first means of verification!
( See the foot-note as an example)
So, in my opinion, as long as they are permitted by law and resources, and as long as they are accurate enough, these data should also be added. It may be up to Google to decide upon which data is made publicly accessible, later.
If anything is obvious at all, then let it be publicly obvious than just for terrorists and sabotages.
These are my independent and personal thoughts. Corroborations and disagreements are welcome. :)
-Viswam
As a semi-off-topic foot-note,
On the day of recent heli-copter crash incident at Nallamallai forests of Andhra Pradesh, India, I had marked up the possible area of crash in a 20 x 5 km strip on both Google Maps and in Wikimapia. The factors like last radio-call time, last sight line report from villagers at Rollapenta and Rudrakonda, the envisaged flight path and salient air-time specs of Bell B-430 with 5 passengers were enough to mark up this data! (All available on public websites).
As more information were getting through, this area was gradually trimmed down to a even narrower and shorter range. By 06.00 Hours UTC next day morning, I was within
less than 5 Kilometers of the actual spot where it happened!These markings are available at
1. http://maps.google.com/maps/ms?msa=0&msid=111493905149591053884.000472a532faed3e3abc1
2. Wikimapia: http://wikimapia.org/#lat=15.8120744&lon=78.7067413&z=13&l=0&m=b&search=nallamalla
(Interestingly, just by coincidence, The 400 KV HT Lines through Srisailam hydel project are just in the vicinity of this crash site!)