Re: Major interview

[Ray Olszewski <ray@comarre.com>]

At 02:42 PM 9/2/99 -0400, Doug Loss wrote [in part]:
>Roman Kirsanow has talked to Liz Coolbaugh from Linux Weekly News about
>doing an interview on the state of Linux in education (more or less).
....
>OK, folks.  What should our approach be in this interview?  What things
>do we want to emphasize, and what things should we stay away from?

An important question to begin with is: will the resulting article be
designed to inform educators about Linux, or Linux folks about education?
Given the publication involved, I'd expect the second emphasis to dominate,
and my comments here are based on that interpretation of the opportunity. 

Linux activists as a group are, to my eye, a lot more realistic about Linux
than they were even a year ago. I no longer see much yahoo boosterism about
Linux ("Linux is perfect and it gets better every day!"); instead, I see
serious assessments of Linux's strengths and weaknesses, coupled with
discussion of how Linux needs to improve if it is to continue to gain market
share -- and especially, if it is to become an important desktop OS. SEUL
(the parent group) is one important player in this last effort. I'd
encourage you to use this opportunity to promote some interest in the
improvements needed to make Linux a first-choice OS for educators, on the
desktop as well as in servers.

First, what are Linux's current strengths in education?

1. Cost -- always a tough piece of the reality of getting technology into
schools. Illustrate how Linux can permit much greater access to technology
than the commercial alternatives because it

        - provides good performance with older equipment in server roles 
                (use LRP as an extreme example of this, but any mainline
                distribution is a moderately good example of it)

        - allows recycling of older equipment into XTerminals, permitting
                a school to buy a small number of powerful, expensive
                servers and share them easily. (Use the case study at
                http://www.silvervalley.k12.ca.us/chobbs/xterms/ as a 
                superb example of this.)

If you can, you might try to pull together some stuff on the numbers of
computers that have been purchased by schools in the past few years, then
consider how few of them are powerful enough to support an upgrade to
Win2000. Schools don't normally replace ANYTHING on the 3-year (or shorter)
cycles that are common to computers in business - this is a looming crisis
that Linux can play a big role in solving.

2. Reliability -- schools generally can't afford high-price tech staffing,
so they need systems that can be maintained by amateurs or part-time
sysadmins (or student sysadmins-in-training). Linux is good for this because it

        - runs reliably once set up, with crashes much less often than
                WinNT and less need to reboot after making small 
                modifications.

        - works well in mixed-platform environments, supporting WinXX and
                Apple file- and print-sharing protocols.

3. Security -- Linux is less vulnerable to damage from the exploratory
fiddling ofunpriveleged users. Also it's less virus prone and inherently
less vulnerable to viruses.

Areas in which Linux needs to improve are:

1. Development of apps that fill specific educational needs. Use the
familiar examples from this list to illustrate what can be done, but be
ready with a bunch of examples of apps that we would like to see available,
things that would give Linux a real edge over Windows in head-to-head
comparisons. This is probably the single most important thing you can get
across in the interview, if you can manage to be sufficiently concrete to
interest some developers (which may require addressing the question of how a
developer can make money developing Open Source applications for schools - a
question I don't know how to answer - I hope you do, or someone on the list
does).

2. Simplified installation. There are several new distributions -- all now
in beta as far as I know (independence, easyLinux, wholeLinux, Corel Linux)
-- that promise to address the need to automate a desktop-oriented Linux
installation. In particular, installation of X needs to be better automated
(why hasn't the XFree development team come up with a way to detect video
cards automatically?).

3. Better availability of standard desktop apps -- basically, word
processing and spreadsheets that can reliable exchange files with MS Word
and Excel. I don't know first hand the quality of the available apps
(WordPerfect; Star Office, Applix), and reports I see on other lists are
mixed. I'll be curious if SUn's current push of Star bears any fruit. The
sorry state of Netscape on Linux (or am I the only person who sees the Linux
version as much less elegant and reliable than the Windows version) also
needs addressing - the Open Source alternatives I've seen in the Debian
distribution are flaky ... and no, lynx is NOT good enough.

4. Inexpensive desktop machines with Linux pre-installed. The leading Linux
hardware companies, like VA Linux Systems, emphasize high-end workstations,
servers, and laptops. Linux needs an easy-to-buy $400 pre-configured
workstation - like an E-Machine. There have been a couple of tries at this,
but I haven't yet seen a successful one. 

Hope these thoughts prove to be of some help to you in your preparations.
Good luck.



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