Re: Major interview

[Ray Olszewski <ray@comarre.com>]

Doug -- this sketch seems very good, covering all the key points. Below,
I've made a few detailed comments, intended to help you flesh out some of
these ideas in the concrete terms that will, I think, make for a better
interview.

At 04:02 PM 9/3/99 -0400, Doug Loss wrote:
>>From reading through this thread I think I see some themes developing.  I'll 
>tell you how it looks to me, and you folks can correct me where I'm wrong.
>
>Linux is usable _now_ in K12 schools in a server setting.  There's nothing 
>technical standing in its way, only perceptions and the common hesitancy to 
>try something new.  I think this is a theme that Jose 
>might be best to advance, 
>with special note being made of OpenClassroom's Local 
>Tutors program and Bill's 
>LENPP. These are ways in which members of the Linux community can get involved 
>in education and support Linux at the same time.

You might point to examples of where this involvement is already occurring.
The Portland and NYC LUGs are the best known, probably because they have
associated majordomo lists. There's been some activity in the Bay Area too,
though I'm embarrassed to say I don't know the details -- it seems mostly to
happen in places that are fairly distant from me, like the far end of the
Easy Bay. Perhaps others on this list can offer specific projects they've
been involved in.

>Linux provides major financial benefits for education over other operating 
>systems, not the least of which is the ability to continue to use computers 
>that can no longer run the supported commercial OSs.  This theme works best as 
>an advocacy tool for those trying to convince 
>curious school administrations to 
>try Linux.  

I think it is important to emphasize this payoff over the "Linux is free"
element. Again, here you can point to specialized distributions and
installers that recycle old computers into XTerminals, the use of relatively
old equipment for single-purpose servers (e.g., mail server, DNS, even
low-volume Web), and the use of these machines with command-line interfaces
to teach C, C++, and Java programming in high schools.

>Along with this goes the theme of open source development, the kind 
>of peer review of work that is very familiar to the educational community.  
>This development model would allow any interested student or teacher to study 
>the internal workings of open source software and tinker with them if they 
>want to.  As someone said, the Linux and educational communities share similar 
>philosophical stances.  These themes are probably mine to address.

This raises a thorny issue. Although teachers accept low pay (relative to
their skill levels and realistic job alternatives), they don't expect to
teach for free. Yet they often expect to be able to get software for free.
There's a good case for the argument that the development of specialized
educational software on *all* platforms is inhibited by the perception that
schools are not a good market, unwilling to pay for good products and often
willing to use software without paying license fees. 

How common these behaviors are, I don't know -- that they exist, I do know
(from my own work experience). But while the maker of a product that has
other markets (a word processor or spreadsheet, say) can let schools use it
cheap or for free, a commercial maker of software designed for use in
schools has no other market to recover its costs from. Consider how many of
the child-oriented products schools have found useful (at least here in the
US), such as KidPix, the Learning Company games, Carmen Sandiego, etc., have
the home market as their main target. I wonder, for example, how much
revenue Topologika derives from schools and how much from home users --
perhaps Roman can say something about this.

I can't tell you what to say here, since what you say needs to be based on
what you believe, not what I believe. So I will only say that you should be
ready to address this issue if it comes up in the interview. 

>For use on the student desktop Linux needs some work.  There currently are few 
>pedagogical or exploration programs available for Linux that would be useful 
>at this level.  While there are open source projects (which we support and 
>encourage more of) currently to address this lack, it seems likely that the 
>lack of these programs will slow Linux's adoption in schools.  

Someone mentioned the role of grants here. Might we (or the Linux community)
want to find a way of encouraging some of the foundations that traditionally
support education technology -- can't think of names right now, but I'd look
at who supports the classroom-oriented stuff that turns up on PBS -- to seed
development of Open Source programs that support acquisition of specific,
non-computer skills? 

Also, what about the role of university-level education programs? Open
Source projects generally reflect the interests of computer-science types
largely, I think, because CS graduate students play a major role in Open
Source development (look at the history of GIMP, apache, Mosaic, for
examples - there are no doubt many more, but these I can name off the top of
my head). What if students pursuing Ph.D and Ed.D degrees (and their non-US
equivalents) were mobilized to develop ed software based on good pedagogical
approaches? What would it take to create an interest comparable to that of
graduate CS students in other types of software?

>We encourage 
>commercial educational software companies to port their existing programs to 
>Linux, and will be happy to discuss with them ways of making a profit with 
>commercial educational software in the Linux world.  

If you were to say this to me in an inteview, I would *immediately* ask you
for some specifics -- what ideas would you suggest? You should be ready with
some.

>We are generally 
>supporters of open source software, but we recognize that that may not be the 
>most appropriate development model for all niches of software.  Roman is 
>the best one for this theme.

Do we support Open Source because it is *free* or because the software is
*better* than commercial equivalents? From the postings of people who fit
this model, there seems to be a mix of the two reasons. The first reason is
a fundamental barrier to attracting commercial developers; the second reason
at least holds open the possibility that if someone were to figure out a way
of making money doing Open Source development (for example, if Cygnus proves
able to make its two-tier model work or if one of these Web sites where
interested parties pre-pay for development succeeds), the education
community would be willing to back up its needs with money.

>We think that much of the software available, both for Linux and for other 
>OSs, is more difficult to use than is necessary and that this is keeping 
>many from trying Linux.  That's one of the reasons we want to see high-quality 
>programs designed for children on Linux--because a user interface simple and 
>obvious enough for children should also be very good for other end users and 
>would help advance SEUL's goals.  

Would it? Why? If I take this thought outside computing for a moment, I
can't think of another example where I would say that adults would benefit
from using a product designed for (say) 10 year olds. In this respect,
"children" is too broad a term -- a UI good for a pre-literate 6 year old
would be maddening for the typical high school student. To take your own
example -- I've used KidPix, and I use Photoshop (closest I can get to GIMP)
-- I would find the limited, cutesy interface approach of KidPix a constant
irritant as I tried to do sophisticated image manipulation. 

There's a sigline I see that quotes Einstein as saying something like:
"Theories should be as simple as possible, but not simpler than that." The
same is true for UIs -- and I think generally that kid-level UIs are too
simple for adults.

>On the other hand, we think that such UIs 
>could and probably should be adopted as alternatives for standard Linux 
>programs such as the GIMP.  We'd like to see kids using the GIMP through an 
>interface similar to Kid Pix, which could be gradually enriched till they were 
>using the full power of the program.  The same goes for word processors, 
>databases, spreadsheets, etc.  This is a theme we can all hit.

Part of this theme is that schools need for their Linux desktops the same
things that *all* desktop users of Linux need, like good word processing and
spreadsheets, a better Web browser, and installation and updates that do not
require specialized expertise. There are also specialized needs of schools,
both pedagogical apps and support apps like gradebooks ... and LAN-level
needs that do require sysadmin expertise ... but included in what schools
need is what *everyone* needs.

------------------------------------"Never tell me the odds!"---
Ray Olszewski                                        -- Han Solo
Palo Alto, CA  94303-3603    	 	        ray@comarre.com        
----------------------------------------------------------------