Re: The Final Word on Browsers and the Future

Robert P Cunningham (bob@lava.net)
Fri, 18 Oct 96 15:23 WET


Message-Id: <m0vEQ83-000A40C@malasada.lava.net>
Date: Fri, 18 Oct 96 15:23 WET
From: bob@lava.net (Robert P Cunningham)
To: www-html@w3.org
Subject: Re: The Final Word on Browsers and the Future

>They make an informed decision not to upgrade -- why would anyone do
>this?   I understand device limitations, but why would a person serious
>about accessing the Internet access it with Netscape 1.0 when 3.0 Gold is
>available?

Since I still get 10%+ hits from Mozilla/0.* and Mozilla/1.* on
several sites, I've made an effort to try to contact some of those
folks over the last couple of months, to ask them that precise
question.  Here's the answers I received (most folks gave more than
one of these reasons), roughly in order from most often to least
often:

	1. No compelling reason to upgrade.  (These folks simply
	aren't attracted to the new features.  Strange but true,
	this is by far the most common thing I heard.)  When pressed,
	most of these folks simply said they didn't think they needed
	anything except a basic no-frills browser to do the kind
	of browsing they do.

	2. Device limitations.  (These folks run systems with
	relatively small amounts of memory and/or disk space, and
	simply don't want to use a browser which consumes more
	machine resources.)

	3. We paid for [typically Navigator 1.*], but don't want
	to keep paying.  (A couple of people who mentioned this
	also claimed that by not buying new versions as each came
	out, they've saved perhaps as much as several hundred
	dollars per machine.).

	4. We didn't pay for [typically Navigator 0.9*], and don't
	particularly want to pay.  (I didn't ask why these folks
	thought the early versions were; but I think they're honest
	people...at least to the extent that if they did use a
	newer version, they'd pay whatever the manufacturer asked.)

	5. We're worried about possible security problems.  (Problems
	mentioned included browser problems, but also potential
	security loopholes in Java, JavaScript.  Quite a few people
	seemed to have reservations about plug-ins which read and
	write local files...as well as just possible browser security
	problems.  And, yes, yes, I know that older browser versions
	have various security problems; these people just seemed to
	prefer the "devil they knew".)

	6. Don't want to be bothered spending the time to keep
	upgrading.  (These folks often seemed both hazy and a little
	defensive about this...it may well be that they're simply
	lazy:-)

	7. Waiting for [some feature].  Similar to the above, but
	these people seemed to be waiting for something quite
	specific.  Most commonly:  math mode, and "ability to
	disable cookies".

	8. Don't like [recent feature].  (Java and JavaScript were
	mentioned, but by far the feature that people most wanted
	to avoid seemed to be frames.  A couple of people specifically
	said they'd stopped using a more current version of Navigator
	and went back to using an earlier version just because they
	didn't like dealing with frames.)

It was an entirely unscientific survey, I only talked to about two
dozen people over the last couple of months.  And I didn't accurately
record the frequencies of all the different answers.  I focused
strictly on people using old versions of Navigator, and didn't
bring up Microsoft IE at all) neither did any of the people I talked
to, which in retrospect seems a little strange, since at least some
of the people I talked to who mentioned money must have know that
it's free).  I talked to a few people running various old versions
of Mosaic (different brands), and while I didn't talk to many of
them, their answers, too, seemed to fall into pretty much the same
categories.

What impressed me is that everyone I talked to actually seemed to
have a reason (one or more of the above) not to upgrade.  No one
said, "now that you mention it, maybe we should upgrade" (although
I supposed I could have convinced a few of them to do so if I'd
tried).  In other words, most actually seemed consciously committed--at
least for now--not to upgrade.


Incidentally, these are pre-2.0 Mozilla browsers I've seen most
often so far this month, roughly in order of most to least often.
The interesting things to note are: 1) how many of them are
"compatible" browsers (not actually Netscape Navigator), 2) how
many are "via proxy gateway" (and thus corporate-supplied browsers
perhaps?), and 3) that, yes, people are still using 0.96, 0.94,
and even 0.93!

Mozilla/1.22 (compatible; MSIE 2.0; Windows 95)
Mozilla/1.22 (Windows; I; 16bit)
Mozilla/1.1N (Macintosh; I; 68K)
Mozilla/1.0N (Windows)
Mozilla/1.2b5 (Windows; I; 16bit)
Mozilla/1.1N (Windows; I; 16bit)
Mozilla/1.22 (compatible; MSIE 1.5; Windows NT)
Mozilla/1.2N (Windows; I; 16bit)
Mozilla/1.22 (Windows; I; 32bit)
Mozilla/1.22ATT (Windows; U; 16bit)
Mozilla/1.22 (compatible; MSIE 2.0;  Windows 3.1)
Mozilla/1.12(Macintosh; I; PPC)
Mozilla/1.12(Macintosh; I; 68K)
Mozilla/1.22 (Windows; U; 16bit)
Mozilla/1.22 (compatible; MSIE 2.0c; Windows 95)
Mozilla/1.22 (Windows; I; 16bit)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/1.22 (compatible; Quarterdeck Mosaic Version 2.02.012 (Mar 23 1996)/Windows/Export)
Mozilla/1.1 (Macintosh; U; 68K)
Mozilla/1.1NOV (Windows; I; 16bit)
Mozilla/1.22KIT (Windows; U; 16bit)
Mozilla/1.1N (Windows; I; 16bit) via Squid Cache version 1.0.2
Mozilla/1.2 (Windows; U; 16bit)
Mozilla/1.22 (compatible; SPRY_Mosaic/v9.20; Windows 16-bit) SPRY_package/v4.10
Mozilla/1.1PE (Windows; I; 16bit)
Mozilla/1.1 (Windows; U; 16bit)
Mozilla/1.22 (compatible; MSIE 2.0B; Mac_68000)
Mozilla/1.0N (Macintosh)
Mozilla/1.2 (compatible; PCN-The PointCast Network 1.1/win16/1)
Mozilla/1.22MII (Windows; I; 16bit)
Mozilla/1.1N (Macintosh; I; PPC)
Mozilla/1.22 (compatible; MSIE 1.5; Windows)
Mozilla/1.12I [ja] (Windows; I; 32bit)
Mozilla/1.1 (Windows; I; 16bit)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/1.1I [ja]  (Macintosh; I; PPC) via Squid Cache version 1.0.0
Mozilla/1.12APPLE (Macintosh; U; 68K)
Mozilla/1.22 (compatible; MSIE 2.0; Windows 95) via Harvest Cache version 2.1-beta-internal-41
Mozilla/1.12 (X11; I; AIX 2)
Mozilla/1.0 (Windows)
Mozilla/1.1 (Macintosh; I; PPC)
Mozilla/1.12(Macintosh; U; 68K)
Mozilla/1.2N (Windows; I; 16bit) via Harvest Cache version 2.1pl1
Mozilla/1.2N (Windows; I; 16bit) via Squid Cache version 1.0.17
Mozilla/0.94 Beta (Windows)
Mozilla/1.2N (Windows; I; 16bit)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/1.12I [ja]  (Macintosh; I; PPC)
Mozilla/1.22 (Windows; I; 16bit) via Squid Cache version 1.0.18
Mozilla/1.22 (compatible; MSIE 2.0; Windows 95)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/1.2 (compatible; PCN-The PointCast Network 1.0/win16)
Mozilla/1.1 (Macintosh; U; PPC)
Mozilla/1.12I [fr]MII (Windows; I; 16bit)
Mozilla/1.1N (X11; I; HP-UX A.09.03 9000/715)
Mozilla/1.1 (Windows; U; 16bit)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/1.22 (compatible; MSIE 2.0; Windows 95) via Harvest Cache version 2.1pl1
Mozilla/1.1N (Windows; I; 16bit)  via proxy gateway  CERN-HTTPD/3.0
Mozilla/1.1I [ja] (Windows; I; 16bit)
Mozilla/1.1BTG (Windows; U; 16bit)
Mozilla/1.12ISUN (X11; I; SunOS 5.4 i86pc)
Mozilla/1.1N (X11; I; SunOS 4.1.3_U1 sun4m)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/0.96 Beta (Windows)
Mozilla/1.22 (compatible; MS FrontPage 1.1)
Mozilla/1.22KIT (Windows; I; 16bit) via Squid Cache version 1.0.16
Mozilla/1.2PE-D (Windows; U; 16bit)
Mozilla/0.93 Beta (Windows)
Mozilla/1.22 (compatible; SPRY_Mosaic/v10.05; Windows 16-bit) SPRY_package/v4.1
Mozilla/1.22 (compatible; MSIE 2.0d; Windows NT)
Mozilla/1.2 (compatible; PCN-The PointCast Network 1.1/win16/1)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17
Mozilla/1.2b5 (Windows; I; 16bit)  via proxy gateway  CERN-HTTPD/3.0 libwww/2.17