License Proliferation

[Robin 'Roblimo' Miller]

>  Does OSI certify all licenses that comply with the OSD?
>
>or
>
>  Does OSI certify all good licenses that comply with the OSD?
>

I believe OSI should certify all licenses that comply with the OSD. 
Otherwise the OSI becomes like a city zoning board or other regulatory 
agency that  doesn't merely  set down requirements for  results  ("All 
windows in new construction should be able to withstand 140 MPH winds or 
be be provided with removable coverings able to withstand 140 MPH 
winds")  but  specifies brands of windows and otherwise micromanges the 
details of even the most minor construction or renovation project.

I also believe OSI should provide a list of "recommended" licenses. This 
would make the OSI into a valued advisor, like a building inspector who 
suggests particular construction methods and materials based on years of 
experience, rather than a bureaucratic overlord that seems to operate by 
whim and and makes decisions based on politics, not suitability for use.

Another possibilbility: Offer a list of "most popular" licenses. 
SourceForge.net could supply this data in multiple forms, if asked. You 
can already get a good sense of license popularity by going to - 
http://sourceforge.net/softwaremap/trove_list.php?form_cat=14 - but it 
might be more useful, and fairer to newer licenses, to show not only 
all-time popularity but also licenses chosen by newer projects over the 
last 30/90/365 days. This would also make trends in licensing apparent, 
which is useful data both individual and corporate open source 
developers should have in hand before they make licensing decisions.

Bottom line: I see the OSI as a standards-setting body, similar to the 
SAE (Society  of Automotive Engineers) or  W3C (World Wide Web 
Consortium).  The SAE sets standards for fastener dimensions, strength, 
corrosion resistance, and other characteristics but doesn't tell a 
production machine shop *how* to achieve those standards. The W3C sets 
standards for valid HTML and other code read by Web browsers, but 
doesn't tell you what server software you should use to deliver that code.

Standards-setting bodies concern themselves with results, not with the 
mechanisms used to meet their standards.

Robin 'Roblimo' Miller
Editor in Chief, OSTG