what matters

[Sameer Verma]

Albert Cahalan wrote:
> It's clear that we aren't all here for the same thing.
> Some wish to help all kids, or poor kids, or non-Western
> kids. Some wish to advance freedom of speech, freedom from
> EULA slavery, or freedom to learn heretical ideas.
>
> Some of us are, assuming good intentions, extremely innocent
> regarding Microsoft. The historical record shows that those
> who partner with Microsoft will be betrayed in the worst way.
> Read "The Scorpion and the Frog" to understand Microsoft.
>
> To a very limited extent, I agree with the idea that we should
> not be pedantic about free software. This does not mean giving
> up the battle in any way. (example: my free wireless firmware)
> Of particular importance is never letting proprietary software
> hold your data hostage, run outside of a sandbox, or become a
> platform (API/ABI) upon which you build.
>
> The critical things I see for a kid's laptop are:
>
> 1. All security-related code is trustworthy. This of course
> means an open source OS, from kernel to web browser. It most
> likely means an open source firmware and boot loader as well.
> Of course, "trustworthy" is to be interpreted from the view
> of the user.
>
> 2. Storage of user data in open formats is easier than storage
> in closed formats. The system must not encourage lock-in.
>
> 3. The user's files are protected. Untrusted programs may
> be used, with the user being certain that they can not secretly
> steal or corrupt his files.
>
> 4. The browser provides strong isolation. A new security partition
> is created when a URL is typed or a bookmark is invoked. A web
> site in one security partition does not get any awareness of any
> other security partitions, EVEN IF THEY INVOLVE THE SAME SITE.
> (Maybe one wishes to have multiple accounts on a web site.)
>
> 5. The browser also strongly isolates plug-ins. It is not OK for
> any random browser plug-in to have full access to the memory or
> files that the browser would have access to. This sort of thing
> needs to be enforced at the kernel level, via the browser causing
> SE Linux to do various role transitions.
>
> 6. The user can use any normal Internet connection, anonymously.
>
> 7. The user can localize. This means more than adding language
> strings to an existing locale. It means creating whole new
> locales and fixing any locale-related bugs which may appear.
> This obviously requires open source software.
>
> 8. There are very few hardware platforms. Ideally there would be
> exactly one, but it must be expected that multiple hardware
> generations will be in use. Each hardware platform gets an optimized
> build that includes hardware-specific image and font sizes.
> It makes a huge difference to the overall experience when all
> the software can assume specific hardware. It's so many little
> things, like the Ruler activity knowing how big to draw itself
> and the Acoustic Measure activity having correct laptop pictures.
>
> In practice, the above pretty much requires GPL-like licensing
> and widespread hardware availability. It doesn't require flash,
> python, a mesh, or even sugar.
> _______________________________________________
> Devel mailing list
> Devel at lists.laptop.org
> http://lists.laptop.org/listinfo/devel
>   
For what its worth, here's something that might help in analyzing the 
situation some more. Its an analytical approach called "mission and core 
competencies (MCC) matrix".

I do not have rights to distribute a copy of the paper, so you'll have 
to look for one in your own libraries. Here's the abstract and a link to 
details.

The MCC decision matrix:: a tool for applying strategic logic to 
everyday activity

Abstract:

Proposes the MCC decision matrix as a modern equivalent of the original 
Boston Consulting Group (BCG) portfolio matrix. Focuses on the key 
questions of mission and core competences which define the axes of the 
matrix and give it its name. Unlike the BCG matrix, the MCC matrix can 
be used to assess the relative merits of any competing claims on 
resources, not just those associated with product/ market segments or 
strategic business units. Also the validity of the matrix itself can be 
challenged and its expression modified. Claims that the matrix can be 
applied to sub-units at any level of the organization and is therefore a 
powerful tool for applying strategic logic to all types of 
resource-allocation decisions throughout an organization, ensuring their 
optimal contribution to building competitive strength. Presents a 
nine-cell directional policy matrix as an alternative form.

http://www.emeraldinsight.com/Insight/viewContentItem.do;jsessionid=4C2BD0F55E3CB5F4C4F15AD390AB6E62?contentType=Article&hdAction=lnkpdf&contentId=864777

Of course, the key dependencies are that you actually *know* what the 
mission and core competencies are.

Sameer

-- 
Dr. Sameer Verma, Ph.D.
Associate Professor of Information Systems
San Francisco State University
San Francisco CA 94132 USA
http://verma.sfsu.edu/
http://opensource.sfsu.edu/