What, if any, is the relationship between user demographics and
browser technology?

Many web sites are specific to browser make/versions (e.g., IE 5+),
configuration options (such as cookies enabled) or plug-ins (such
as Flash).  The companies who build such sites say that they're
designing for a target audience described in terms such as 17-30yr
old males in North America.  Is there actually any correlation
between gender, age, location, income (that is, typical marketing
demographics) and browser technology?

I'd like to know of research or survey evidence (preferably not
more than 2 years old) rather than anecdotal evidence.  If the
evidence is from a web-based survey, the survey itself should
have been compatible with any browser set-up (that is, it didn't
depend on things like images, cookies or JavaScript) which rules
out anything from StatMarket and similar site stats aggregators.

Suggested places to look:

- (browsers) http://www.nua.ie/surveys/index.cgi?f=FS&cat_id=5
- (demographics) http://www.nua.ie/surveys/index.cgi?f=FS&cat_id=18
- (usage patterns) http://www.nua.ie/surveys/index.cgi?f=FS&cat_id=34
- http://www.useit.com/
- http://www.acm.org/archives/wa.cgi?S1=chi-web (CHI-WEB archive)

Surveys already found:

- GVU's 10th Survey (1998)
  http://www.gvu.gatech.edu/user_surveys/survey-1998-10/graphs/technology/q41.htm

- "No More Browser War Say Zona Research" (1999)
  http://www.nua.ie/surveys/?f=VS&art_id=905355396

---

Request for Question Clarification by larre-ga on 31 May 2002 16:53 PDT

Hello, 
 
I'm familiar with of the sources of statistics you've listed,
including the archives going back several years. They do contain
fragments of data, but are really too scattered to compile into a
meaningful report, since they are based on a variety of different
samples. Other public sector sources of hard data are extremely rare,
though anecdotal and speculative resources may be located. To obtain
correlated statistics, you'll probably need to purchase them.
Proprietary terms of use closely govern the ability to make these
statistics public by those who purchase the reports. The companies who
compile them actively defend their intellectual properties and source
reporting methods. For the most part, you'll find correlated
statistics in eCommerce type reports, which datamine customer
purchasing statistics matched to server usage.
 
Selected Report Sources: 
 
Jupiter Media Metrix 
<a href="http://www.jmm.com/index.html">http://www.jmm.com/index.html</a> 
Market Forecast Report 
<a href="http://commerce.jmm.com/jupdirect.asp?mode=display&amp;itemid=1905">http://commerce.jmm.com/jupdirect.asp?mode=display&amp;itemid=1905</a> 
 
Nielsen//Net Ratings 
<a href="http://www.nielsen-netratings.com">http://www.nielsen-netratings.com</a> 
E-Commerce Intelligence 
<a href="http://www.nielsen-netratings.com/marketing/ecommerce/ecommerce.htm">http://www.nielsen-netratings.com/marketing/ecommerce/ecommerce.htm</a> 
 
All Net Research - a division of Internet.com 
<a href="http://allnetresearch.internet.com">http://allnetresearch.internet.com</a> 
 
E-Marketer Stat Database 
Monthly or Yearly Subscription 
<a href="http://www2.emarketer.com/">http://www2.emarketer.com/</a> 
 
Forrester Research 
<a href="http://www.forrester.com">http://www.forrester.com</a> 
Flash and the Executable Net 
<a href="http://www.forrester.com/ER/Research/Brief/Excerpt/0,1317,14852,00.html">http://www.forrester.com/ER/Research/Brief/Excerpt/0,1317,14852,00.html</a>
Guest Access: 
<a href="http://www.forrester.com/Email/Global/1,5735,0,00.html">http://www.forrester.com/Email/Global/1,5735,0,00.html</a> 
 
Two do-it-yourself survey methods might help you obtain hard numbers
for a relatively small internet population sample. You might approach
selected eCommerce sites or their developers, and inquire if their
site statistics might be available for use in a White Paper. Or
approach the owner of a well-managed mailing list. Many list owners
have detailed demographics about their subscribers, as well as server
statistics from their web signup pages.
 
larre-ga 

---

Clarification of Question by paola-ga on 01 Jun 2002 05:59 PDT

larre, what clarification do you need from me? 
 
That there wasn't any readily-available evidence to support the 
relationship goes most of the way to answering my underlying 
question.  You're welcome to respond to the rest of this in an 
Answer so that we can complete this transaction. 
 
Here's what prompted me to ask the question now: I just read this 
quote from a web designer in the June 2002 issue of Cre@te Online 
(a UK Internet magazine): "with entertainment stuff, we're starting 
to rule out people who aren't Flash-based and who don't have a 
fast connection. [...]  We'd love to do more HTML stuff and give 
the experience to everyone, but inevitably it's 18 to 35 ABC1 and 
that's it." 
 
(I don't know if the social class codes are universal - ABC1 
describes 68% of the population and includes everyone except 
manual workers and people dependent on the state: 
<a href="http://www.icmresearch.co.uk/research%5Fquestions%5Fanswered/what_is_social_class.htm">http://www.icmresearch.co.uk/research%5Fquestions%5Fanswered/what_is_social_class.htm</a>)
 
I want to know how people get from &quot;18-35 ABC1&quot; to&quot; Flash-based 
with fast connections&quot;. 
 
Amongst the many other reasons to design browser-specific web sites, 
I think that money is likely the real main factor; it must be easier 
to build something browser-specific but I get the impression that 
companies charge more for, say, a Flash-based site.  Some developers 
mistakenly use stats based on their own or aggregated server logs 
from browser-specific sites to justify their continued focus on sites 
specific to browsers, platforms, plug-ins, technologies, whatever. 
 
I will mail the research companies you listed to see if they have 
any reports with the correlated data I'm interested in but, having 
now spent a couple of hours trawling through their report lists, 
I suspect that such correlated data doesn't actually exist. 
 
Thanks for the suggestion to contact eCommerce site owners and for 
contacting list maintainers.  I administer a 300-member mailing list; 
members are mostly academics, so I would only be able to correlate 
browser usage with gender and age.  I also have access to seven years 
of web server logs from a variety of web sites - the trick would 
be approaching the site which would appeal to all social classes. 
 
 
Paola 

---

Request for Question Clarification by xemion-ga on 01 Jun 2002 15:25 PDT

Can you better define "browser technology"?  What do you mean by that? 
 
Are you looking for a survey that says something like this: 
Ages 12-18 - IE6.0 40%, IE.50 20%, NS4.7 40% and so on... 
 
Thanks. 
 
xemion-ga 

I think I have the information you are looking for but first some
2001/2002 statistics:

86% uses IE
http://www.nua.ie/surveys/index.cgi?f=VS&art_id=905355869&rel=true


2002 

http://cyberatlas.internet.com/big_picture/geographics/article/0,,5911_1011491,00.html#table
66% of US adult population use the net. Biggest users -18-29 old
males.

2001
http://cyberatlas.internet.com/big_picture/geographics/article/0,1323,5911_919221,00.html.
"Harris found the online population still skewed toward the more
educated, more affluent and the non-elderly."  But this is slowly
changing... "But 19 percent of those online have household incomes of
$25,000 or less (compared to 25 percent of the total U.S. population),
38 percent have never been to college (52 percent of U.S. adult
population) and 7 percent are over the age

of 65 (16 percent U.S. adult population). 

2002 http://www.nua.ie/surveys/index.cgi?f=VS&art_id=905357989&rel=true
59 percent of the net is non-english speaking.

2001 http://wcp.oclc.org/ 47% of all website are US based. Germany and
Canada come in second and trird at 5%. & 4%.

2002 - New survey from AOL 
http://www.nua.ie/surveys/index.cgi?f=VS&art_id=905357613&rel=true
states that teenager prefer the internet to the telephone.

2001 http://www.nua.ie/surveys/index.cgi?f=VS&art_id=905357202&rel=true
This survey shows that Students drove Net traffic up in August by 45%.

Net Developers aim their code to these browsers " # Browsers that meet
standards well are IE 5, IE6, Mozilla, Netscape 6, Opera 5, and Opera
6. "
http://www.upsdell.com/BrowserNews/index.htm

I found no specific surveys keying browser use to affluence per se but
Apple users are the exception as they are very loyal to the brand even
though Apple only has a 2.5 market share.
http://www.dailymac.com/articles/opinion/wwdcimac.html

Website designers do pay attention to demographcis as seen in this
article coming from http://www.dreamink.com/design5.shtml. It's a good
one because of these two paragraphs. - "It depends upon which study
you read. Our research indicates that a study of browser usage based
upon a valid sample just doesn't exist. That said, the race, in our
now reasonably well informed opinion, is leaning in favor of MSIE.
(The AOL browser is often counted as MSIE by statisticians, which
accounts for many of the discrepancies in numbers by various
organizations.)

BUT... it doesn't matter. What DOES matter is that there are differing
browsers with differing characteristics, and there are LOTS of them
already out there. Even if the best "selling" browser changes, the
chances of the majority of users converting to that browser within a
short period of time are very slim. Even if the percentage of users
drops lower, 25% of 276+ million is a respectable figure - around 80
million. Both Internet Explorer and Netscape browsers have their
advocates. They'll both be around for some time to come.

Particular emphasis is given on compatiablily because - "DESIGN
IMPACT: Use of the Netscape 216 Color Palette is highly recommended.
If coding under the HTML 4.0 Standard the "Transitional" format should
be selected. All pages should be viewed and tested in both MSIE and
Netscape browsers. If browser specific design elements are chosen,
then a Javascript to detect and direct visitors to proper versions
becomes a necessity."

(I do this drill very religiously because I am a web developer.)

This last article points out that developer pay close attention to the
Net and net protocols while continuing to use tools specific to the
platform on which their application runs. This is supported by sites
like www.webdeveloper.com and http://hotwired.lycos.com/webmonkey/
that talk trends about new code that runs on the net as well as on
Windows and linux.

MS will always be a player as seen in an ariticle written by Michael
Pastore where - "NT (is) Gaining Ground in Server Market," The article
goes on to say that Linux and Apache are making headway as well..
NThttp://cyberatlas.internet.com/big_picture/hardware/article/0,,5921_157721,00.html


Last but not least is http://www.w3.org the web standards site and
/http://archive.webstandards.org/ a watch dog environment that points
out the best ways to adhere to web standards.

Of course we have not touched on Java and open source but www.sun.com
and http://www.opensource.org/ are great places to start.

HTH

Bob

---

Request for Answer Clarification by paola-ga on 02 Jun 2002 10:41 PDT

Bob, 
 
Sorry, but I think you've answered a question that I didn't ask. 
 
And I specifically said that I wasn't interested in anything based on 
StatMarket data (your first link). 
 
It's one thing to be able to say "most people use browser X" (which 
is what I think your answer addressed).  It's another to say that 
"most male teenagers use browser Y whilst most older retired women 
use browser Z." 
 
I asked what, if any, is the relationship between visitor 
demographics and their browser technology. 
 
By visitor demographics I mean the ways web companies describe 
their target audience (e.g., by social class, age and gender); 
by browser technology I mean things like browser make and version, 
configuration (such as cookies enabled) and plug-ins (such as Flash). 
 
So, I want to know about surveys/reports which cross-reference 
information about visitors against information about their browsers. 
The 1998 GVU 10th survey seems to show that browser choice wasn't 
related to gender or age.  Has such a survey been repeated recently? 
 
To repeat the magazine quote again, a web designer said: "with 
entertainment stuff, we're starting to rule out people who aren't 
Flash-based and who don't have a fast connection. [...]  We'd love 
to do more HTML stuff and give the experience to everyone, but 
inevitably it's 18 to 35 ABC1 and that's it." 
 
Such statements aren't uncommon and, as I said in my clarification, 
I want to know how people get from "18-35 ABC1" to" Flash-based with 
fast connections."  I don't think they can justify the statement - I think 
it's just wishful thinking. 
 
If it helps, I am pretty confident that the answer to my question is 
"none".  I just want some concrete evidence so that when a web 
developer says something like "our web site excludes Mac users 
because our target audience doesn't use Macs" I can then say 
indignantly "how can you possibly know that?" whilst slapping them 
with a wet fish... 
 
 
Paola  

---

Clarification of Answer by remoran-ga on 03 Jun 2002 07:25 PDT

The IE stat is pretty consistent as per 
&lt;a href=&quot;<a href="http://www.upsdell.com/BrowserNews/stat.htm">http://www.upsdell.com/BrowserNews/stat.htm</a>&quot;&gt;<a href="http://www.upsdell.com/BrowserNews/stat.htm">http://www.upsdell.com/BrowserNews/stat.htm</a>&lt;/a&gt;. The only difference is 
that Browser News uses three different sources to get a more accurate 
read on this. I completely agree with Browser news on their assesment 
on how stats can lie. &amp;quot;Caution: statistics can mislead. Caching 
distorts raw data; each site attracts different audiences, with 
different demographics; each survey uses different methodologies; many 
surveys mis-identify certain browsers. Statistics like these, 
therefore, should only be used as very rough predictors. Consider, for 
example, the statistics here, reported by three different sources: 
  
    * Source 1 stats are from sites that use a hit counter, which 
excludes many popular, professionally-made sites. 
    * Source 2 stats, from EWS has browser stats of its visitors. Note 
that it counts hosts rather than page accesses, which skews the stats. 
    * Source 3 stats are from this domain (&lt;a href=&quot;<a href="http://www.upsdell.com">http://www.upsdell.com</a>&quot;&gt;<a href="http://www.upsdell.com">www.upsdell.com</a>&lt;/a&gt;): because 
of its special audience, its stats apply to a narrow segment of the 
population. Note the anomalously high 'other' percentage: the stats 
report program (WebTrends) is likely doing a poor job of identifying 
browsers. 
  
The best stats for a site are the stats gathered for that particular 
site: and even these are skewed by caching and faulty 
browser-detection.&amp;quot; 
  
As a program designer and web developer, the first thing one does is 
to define what kind of program do you want to develop. If it's a game, 
you check out the competition and see if you can improve upon what is 
out there and to make sure it will run on the target platform it is 
suppose to run on. If it's a utility, you use the set of tools that 
works with the environment you are in. (Windows, Linux, Apple etc.) If 
you are building a web app, you look for compatabilily with defacto 
and w3 accepted standards. (Flash, Acrobat - defacto, CSS, XML - W3.) 
From there, you build your app to meet the needs of your target 
audience. (deomgraphis plays the role here. As for browsers, NS 4.7, 
IE5, IE 5.5, NS6, Opera and Mozilla 1 R3 are the major ones that 
adhere to web standards. If you don't test for at least NS 4.7, 
Mozilla and IE5, you will be toast. 
  
The other thing to considere is people (general users here) rarely 
change browsers, particulary if your machine comes with one already 
installed. (See MS and Netscape). If the thing works right, leave it 
alone or upgrade the one you have because &amp;quot;Computers are enough of a 
hassle as it is.&amp;quot;. I have never seen a demographic keyed to browser 
because the major ones all do the same thing. 
  
As for Cookis &amp;quot;Cookies are a general mechanism which server side 
connections (such as CGI scripts) can use to both store and retrieve 
information on the client side of the connection. The addition of a 
simple, persistent, client-side state significantly extends the 
capabilities of Web-based client/server 
applications.&lt;a href=&quot;<a href="http://wp.netscape.com/newsref/std/cookie_spec.html">http://wp.netscape.com/newsref/std/cookie_spec.html</a>&quot;&gt;<a href="http://wp.netscape.com/newsref/std/cookie_spec.html">http://wp.netscape.com/newsref/std/cookie_spec.html</a>&lt;/a&gt; 
  
Microsoft has the same kind of blurb here. 
&lt;a href=&quot;<a href="http://www.microsoft.com/info/cookies.htm">http://www.microsoft.com/info/cookies.htm</a>&quot;&gt;<a href="http://www.microsoft.com/info/cookies.htm">http://www.microsoft.com/info/cookies.htm</a>&lt;/a&gt; 
  
Cookies 2 - &lt;a href=&quot;<a href="http://ciac.llnl.gov/ciac/bulletins/i-034.shtml">http://ciac.llnl.gov/ciac/bulletins/i-034.shtml</a>&quot;&gt;<a href="http://ciac.llnl.gov/ciac/bulletins/i-034.shtml">http://ciac.llnl.gov/ciac/bulletins/i-034.shtml</a>&lt;/a&gt; - This 
defines exactly what a cookie is in terms of a web standard. PROBLEM:  
 Cookies are short pieces of data used by web servers to help 
           identify web users. The popular concepts and rumors about 
what a 
           cookie can do has reached almost mystical proportions,   
           frightening users and worrying their managers. 
PLATFORM:  Any platform that can use a modern web browser. 
DAMAGE:    No damage to files or systems. Cookies are only used to 
identify 
           a web user though they may be used to track a user's 
browsing 
           habits. 
SOLUTION:  No files are destroyed or compromised by cookies, but if   
           you are concerned about being identified or about having 
your web 
           browsing traced through the use of a cookie, set your 
browser to 
           not accept cookies or use one of the new cookie blocking   
           packages. Note that blocking all cookies prevents some 
online 
           services from working. Also, preventing your browser from   
           accepting cookies does not make you an anonymous user, it 
just 
           makes it more difficult to track your usage. &amp;quot;  
  
All modern browsers support cookies and NS 4, IE 5-6, Mozilla and 
Opera, among others, allow users to control the use of cookies without 
a problem. As for Flash and PDF, these browser support them as well. 
Ditto for Flash (also PDF). 
  
As a rejoinder, if a web developer makes the kind of statement similar 
to the one in your request, checks out another web developer because 
that kind of statement is simply not valid. If the app runs only on 
Windows or on Macs and is NOT Net enabled, the WD's statement is 
valid, if the app IS net based -  Not. 
  
As for fast connections (broadband is definitely coming), the approach 
many developers take, particularly if there are intense graphics 
involved, is to do a dual use website with braodband and dialup 
factored in where animated Flash goes the BB route and still imagery 
or no imagery applies for dialup folks. 
  
Hope this clarifies things.   
  
Bob  

The researcher answered a question that I didn't ask and
started with a source that I'd specifically said I wasn't
interested in.  This was after another researcher had already
posted a question clarification because they'd found that there
was no freely-available information answering my question.

I requested an answer clarification from the researcher who posted
the Answer - the result was more answers to different questions and
the phrase "I have never seen [data answering your question]".
The wording implied he'd not researched the question I'd asked when
preparing his Answer; I asked whether this was the case (4 days ago)
and he hasn't responded.

Since larre-ga, the first researcher who responded (but not in a
posted Answer), has provided me with useful information, I'd actually
prefer that the fee goes to larre-ga (and Google) since the service
in itself was useful.

I was provided with an answer to a different question to the one I
actually asked, and again after a question clarification.  I offered a
second chance but received no response.

Bob, thanks for taking the time to respond to my answer clarification
but I still feel that you've not answered the question I asked.

I didn't ask about general browser stats or their validity.

You wrote:
> I have never seen a demographic keyed to browser because the major
> ones all do the same thing."

Did you look for any when researching the answer to my question?

> if a web developer makes the kind of statement similar to the one
> in your request, checks out another web developer because that kind
> of statement is simply not valid.

Where's the research or survey evidence to say that it's not valid?

(Note my initial proviso that I'm only interested in evidence in
which the data-collection method itself was compatible with any
browser set-up, which still rules out StatMarket.)


Paola

We researchers sometimes use the Clarification mechanism to provide
greater definition for questions that, in our individual opinions, are
outside the scope of what we're able to provide. I am a subscriber to
the report services offered. I know that the stats you're looking for
available, however, I can't "resell" those statistics to you.

You're asking if there's still a market for straight HTML based web
development, is that correct? Many of Jakob Nielsen's AlerbBox
articles indirectly address this trend.

The End of Homemade Websites:
http://www.useit.com/alertbox/20011014.html
The End of Web Design
http://www.useit.com/alertbox/20000723.html
Stuck With Old Browsers
http://www.useit.com/alertbox/990418.html

I realize that you might already be familiar with these sources. 

Just a note, I keep the sort of correlated statistics you're seeking
for my own website (200K+ visitors per year), and several eCommerce
websites that I administer, however they are not directly for sale.
They're used only as the basis for custom usability studies
commissioned by potential entrants into niche markets. Such statistics
are a business commodity for many other entities as well, which is why
you'll see high prices for purchase of compiled reports.

~larre-ga

Another interesting response, larre - thank you.

No, I'm not asking whether there's a market for straight HTML-based
web sites.  My question's about browser compatibility.

If someone specifically says "our site only works with x,y,x browsers
because that's what our target audience use" (where "target audience"
must mean a sub-set of the web population) I want to know whether such
a position can be justified by research.

I appreciate that you can't pass on bought or private data. How about
a yes/no answer to a version of my original question?

Is there any research which says that people who use browser x are
significantly different to people who use browser y?

For example, is there a significant difference between users of
Netscape 4 and users of IE 5?  Or between people who disable cookies
and those who don't or those who have the latest version of Flash
and those who have an older version?  (Users defined by their age,
gender, social class and not whether they are disabled.)


Paola

Specifically, Paola, you'll find the differences in those areas to be
more closely related to two factors: level of education, and how long
the user has used the Internet, rather than age or income levels.
There are minor blips in standard demographic categories, but I
wouldn't call them statistically significant. Sample of 1,200 users.
Method: User poll, June 2001.  Website category, Internet,
Subcategory, Education. Average user experience on Internet: 14
months, Median 12 months,  Average age: 28, Median age 29.

Percentage of users who disable cookies 10.8% of total users

Of those users:
0-3 months of Internet usage: 0.5
3-6 months of Internet usage: 4%
6-12 months of Internet usage: 24%
12-24 months of Internet usage: 42%
24-48 months of Internet usage: 20%
48-60 months of Internet usage: 6%
60+ months of Internet usage: 3.5%

Grade 12 or under: 4%
High School grad: 21%
Some college 31%
College grad 22%
Post grad studies 14%
Post grad degree 8%

Whereas, using the same sample, the age breakdown is pretty much
'even'
Under 18 12%
18-25 19%
25-35 18%
35-45 19%
45-60 18%
60+ 14%

By gender:
Female 46%
Male 54%

Here's some supporting info, from roughly the same era, from
NewsBytes. I've had to use Google's cache function, because my
bookmark defaults to the current page/issue. I've fed the long link
into make a shorter link for ease of use.

http://makeashorterlink.com/?Y2D436DF

My experience with browsers vs. demographics shows that the most
significant factor influencing which browser used is the one that was
used the first time the user accessed the Internet. Simple as that.
90% of users do not switch browsers. They will upgrade within the same
browser family, but both actions (switching browsers and speed of
upgrade vs. release date) are most closely related to level of
education. Secondary factor is time online. And Netscape users (all
versions)  averaged 2.9+ years of Internet experience, nearly double
that of IE users (all versions), who averaged 1.6 years Internet
experience. I can't squeeze any other major differences out of the raw
data.

This isn't a complete answer to your question, but a comment on what I
feel are the most significant areas to focus on. I hope this
information offers you a few new directions to explore.

~larre

I have found this an interesting exercise if only because I'm
still getting answers to *different* questions despite already
having got an answer I said I was happy with (in larre's question
clarification).

I already know that computer and Internet experience are related
to choice of technology; the Zona research link in my original
question was about browsers used by IT professionals.

Computer or Internet experience is only relevant to my question when,
say, "IT professionals" are identified as the target audience.
I find it hard to believe that a marketing planning group for a
new web site to sell wine, for example, would define their target
audience in terms of how long web users have used the Internet.
Instead, issues like delivery will imply a geographical location,
and the products being sold will imply things like age, income
and lifestyle.

That said, I was in a sales meeting a while back for a company who
wanted to put their flower-delivery service online.  We talked about
how we build compatible web sites as a matter of course, that their
site would also work with text-only browsers.  Their response was
along the lines of "don't bother with any extra effort - we don't
want geeks or anoraks using our site."

On the chi-web mailing list, I have seen a similar attitude from
other developers: some people justify their incompatible web sites by
specifically saying that they're designing for their target audience.
I assume that "target audience" doesn't mean "any web user"
(when general browser stats would apply) and, by the discussion
context, that "target audience" is defined in marketing terms,
not in technology terms (which, I contend, are completely arbitrary).

In the context I've described, larre basically confirmed that
nothing meaningful can be said about a potential web site visitor
by the browser (or related technology) they use.

I'm going to mail the guy I quoted to see what evidence they have;
I'm also considering mailing the companies that sell reports to see
whether they have the kind of correlated data I'm interested in.
I'll follow-up here - alternatively, anyone interested can mail me
(I'm listed second in a Google search).

Paola

Hi paola

Just as a followup to the answer & clarification I provided.

Demographics on users, as you well know, are very sophisticated and
developers of specific web applications use them religiously to help
insure that their programs will be successful in the marketplace. (net
usage, education, age, etc., etc.) In terms of programming, the key
here is that any application developed must run on all browsers that
work with accepted internet protocols. (IE 5, Opera, Mozilla, NS6  are
the main ones here.) If this is not done, the program in question will
fail. I know this from experience as I am a developer and have gone
through the required steps to make sure my programs do just that.

Regarding the validity issue on the web developer's statement, I will
stand by that because developers can pick and choose from a wide
variety of options (Flash, PDF, CSS etc., etc.,) that are web savvy
and will run on all of the main browsers without a problem. I like
this because it eliminates self imposed restrictions as to what
program will run on what platform. This is why the Net has
revolutionized programming and has turned computing into a universal
enterprise that can be uses by all.

As refererence, here are sites for browser characteristics geared to
web developers. These are very important resources for people like
myself.

http://www.upsdell.com/BrowserNews/browsers.htm is a terrifc site for
browser stats. This url talks of browser futures.

http://www.netmechanic.com/browser-photo/tutorial.htm is a "what to
look for" in difference among the main line browsers. The key here is
not being on the "bleeding edge" and
http://www.alistapart.com/stories/tohell/ talks about staying current
with new browsers that adhere to standards. The majors are promonently
listed.

Bob

PS, Comments made by everyone on your question are absolutely terrific
(wish I had thought about Nielsen).