Mike,
>I haven't called BSA and asked them for there methodolgies, but when I posted that, I asked if someone had different numbers that we could compare. Haven't seen any yet, so far BSA's is the only hint (I didn't say facts just for you ;-) to go off of. You can dismiss it if you want, but I'm not sure you can make any assumptions with absolutlely no numbers.
But you have to understand that at this point in my life...(don't worry, I am not going to make fun of your age since I am not an old man by any means...) I don't much believe _any_ numbers. If I were to find some semi-reputable web site and post numbers off it, I would feel like a hypocrite (since I just made a posting saying these BSA numbers can't be used if you don't know from whence they came...)
So I tend to base my thoughts on my own experiences and my own sense of "common sense" (which falters at times, but I have to try to trust _something_).
Now, I did find this:
http://www.spa.org/piracy/pubs/98g.aspIt is SPA's study for 1998 and it shows methodology. Here is rough breakdown of how it is done:
- The estimates are based mainly on two sets of data...installed software (estimated) and licenses sales (estimated). The disparity between the two is assumed to be based on piracy. [I am with them so far (estimated)]
- Determining installed software is almost COMPLETELY a statistical, trend-based estimate. They start by determining the number of PCs in a country, and then they decide what software is likely to be installed on those PCs. Here is a blurb from the above link:
---------------------------------------------------------------
The level of the PC installed base is a general measure of the level of technological development within a country. But because the number of PCs alone may be misleading, IPR developed for each country a ratio of the installed base of PCs to the number of white-collar workers in that country. U.S. historical trends were also used to estimate the effects of lagged technological development by maturity class. For all countries, each of the four PC classes was assigned to one of five maturity classes based on this ratio.
From market research provided by SPA and BSA member companies, IPR determined the number of software applications installed per PC shipped and then developed software-per-PC ratios for the shipment groups adjusted for the software maturity class of the market.
Third, IPR took into account the type of business software, because piracy rates vary among application types. Grouping the software applications into three tiers and using specific ratios for each tier further refined the ratios. The tiers used were General Productivity Applications, Professional Applications and Utilities. The components within each of these tiers are listed on the next page. These tiers represent different target markets, different price levels and, it is believed, different piracy rates.
---------------------------------------------------------------
- Determining licenses copies is easy...go to the various vendors and get numbers from them (though it wasn't always done like this)...here is a blurb from the above link:
---------------------------------------------------------------
Once IPR estimated the size of the applications in use within a country, they company determined the number of applications that had been legally shipped to that market. For the 1995 and 1996 piracy studies, the main source of data for software shipments was SPA's Data Program. However, that program ceased operation in early 1997. Therefore, the major challenge for the 1997 study was to replace that data source.
IPR's approach was to utilize the member companies of BSA and SPA to develop piracy study sponsors who would volunteer their proprietary shipment data to the study. These data were provided under nondisclosure agreements for the purpose of constructing an accurate estimate of the software industry's 1997 shipments.
---------------------------------------------------------------
That's the gist of it, though you can certainly read more for yourself.
You can guess my opinion...it sounds like a load of hooey. They use the word 'ratio' and 'trend' umpteen times, and some of their correlations seem to be pulled out of thin air (white-collar-to-PCs ratio? How are they defining "white-collar workers"?). Finally, their estimates lack one crucial element (at least no one has presented it), and that is an error range. a study without error bars is absolutely MEANINGLESS. My physics degree taught me nothing if not that...
This all seems strange to me when surely the BSA has been busting folks for a few years now, so why don't they use _real_ figures of how much software is truly running unlicensed? I see lots of figures like "2 out of 4 business have unlicensed software". Great. How much? One copy? Give me _numbers_. Could it be that they don't give full blown numbers because the numbers are smaller than previous estimates? I also see a lot of "Over 2 million dollars in fines..." That's great too, but fines for piracy are very stiff (once you get caught), so those large sums may only be related to few thousand dollars worth of lost software revenue...
Anyway, I better stop...this is getting long. If you want, I will find numbers that are different (lower) than the BSA's estimates, but if the lower estimates were calculated using voodoo like that delineated above, I'd rather not bother -- I wouldn't trust _either_ set of numbers.
JoeK
