html, webtv, etc. (fwd)

Subir Grewal (subir@crl.com)
Wed, 12 Feb 1997 09:38:11 -0800 (PST)


Date: Wed, 12 Feb 1997 09:38:11 -0800 (PST)
From: Subir Grewal <subir@crl.com>
To: HTML Discussion List <www-html@w3.org>
Cc: David Heller <dheller@tisny.com>
Subject: html, webtv, etc. (fwd)
Message-Id: <Pine.SUN.3.95.970212093305.1906B-100000@crl4.crl.com>


Well this message landed in my box earlier today.  It wasn't sent by the
author David Heller <dheller@tisny.com>, whom I don't know.  I thought
it might interest some of the people on this list because we've been
discussing similar issues here.  My take on the whole issue is very
different from the author's since I think the true promise of the web is
as a cross-referenced information system. 

---------- Forwarded message ----------
Date: Wed, 12 Feb 1997 10:27:15 +0000
Subject: html, webtv, etc.

IN A CALL TO ARMS to New York's developer community, guest columnist David
Heller calls for the creation of a Web union to take aim at Microsoft,
Netscape, and W3C. Heller is senior developer and designer in the Human
Active division at Transaction Information Systems, Inc., a systems
integration and application development company.


It's Time for Browser War Victims to Fight Back

So the browser war is big news, and not just in the trades anymore. The New
York Times, CNN, and other mainstream media are now covering it in detail.
But most of this coverage is concerned with market share, stock trading, and
anti-trust laws. It ignores the war's true victims: the developer community.
For the people who craft what everyone sees on the World Wide Web, it's only
going to get worse.

There are two basic problems. The first one, which is well understood, deals
directly with the HTML (Hyper Text Markup Language) specification. The
browsers simultaneously create new "proprietary" HTML tags that the other
can't use. This makes it difficult if not impossible to make Web pages that
work well with both browsers. This increases development time, limits
technologies to lowest common denominators, and often limits market
viewership to a single browser.

The second problem is this: browsers are implementing the same technology,
but are using different language syntax in order to create it. Each claims
to have the support of the open standard body, so its syntax is correct. The
body that supposedly controls this specification, the W3C (The World Wide
Web Consortium) is either a puppet, or ineffectual. Netscape doesn't care
about it any more, because it can just use its market share strength to pull
its proprietary tags through the W3C at a later date. And Microsoft has
somehow gained control over the W3C so that its proposals seem to always go
through. Basically, there is no W3C worth a hoot.

The other problem is that even when the HTML code and syntax is the same,
the way a browser interprets that HTML may be different. HTML is not a
programming language but is an interpreted language. It is up to the browser
to decide how to lay out a page according to the code provided. Sometimes
that can mean that one browser will create a double space, and the other
will single space a paragraph break. That's pretty harmless. But it can also
mean that a table cell has a completely different width, making the same
page look completely different depending on which browsers it is using. The
W3C doesn't coordinate browser end results of HTML but only what HTML they
should be prepared to interpret.

Along Silicon Alley and anywhere else where developers rely on HTML, the
headache is only getting more intense. Creating tables is a nightmare.
Microsoft Internet Explorer and Netscape Navigator implement some basic
interpretations of tables differently. You know what that means: pages with
the same table with cells of various widths, border colors, and even text
layout. PCTV only makes it worse. WebTV, Pippin, and other TV-based Web
browser systems hit the scene and have different rules for displaying data
and different polemics that need to be considered when designing them. For
example, WebTV has decided that frames are bad because of focus group
results. Also, WebTV doesn't support all table functions, because it needs
to consider that its format is on a TV screen, which has a different aspect
ratio, and pages have to fill an entire screen every time (no horizontal
scrolling).

The reason I mention WebTV and its counterparts is because if the Web is
going to fulfill its great potential as a medium for the masses it will to
have to be regulated the way television is. I'm not saying that the Web
should become TV, but its standardization needs to be invented if we are
going to gain all that the Web has to offer. Even VCR makers had to
eventually choose one open system (VHS). The same will be true for the Web.

In the end, clients and end-users are affected by the lack of
standardization in the industry. Development costs will fall on the
end-user, end-users will face ever-growing confusion, and the promise of
certain technologies will never be achieved. Here's an example that hits
home: I am working on a site now that is perfect for a certain new
technology--because the functionality gained is so compelling that not to
use it would be detrimental to the goals of our client. But the technology
in question is not accessible to much of our corporate client base because
of firewalls. So we are forced to build two versions of the site. In order
to keep what's left of the functionality we can do without the new
technology, thus doubling our costs. In the end, the cost will fall on the
client.

In the Web's old days, most advancement in HTML occurred in consideration of
the developer/designer community. There was an outcry from developers and
Netscape took its market share power to the people, and said, "Sure! Here ya
go!" Today, Microsoft is in the picture and you can't do anything that
doesn't work on both browsers, even though they each offer some great
extensions to HTML. And ego prevents either software maker from giving in to
the other.

What I suggest is this: creating a new body to respond to these issues. No
software maker or any company with a single software maker partnership will
belong. The members of this organization would be Web content developers:
HTML coders, systems developers, multimedia experts, and Web application
developers. As a union of sorts, we can "order" the software developers to
correct their system. We choose the best of each browser, and make additions
in an organized fashion.

As a union, this body of developers will wield great influence over the
software companies. It is we who have the greatest contact with our
end-users. We know our audience and can best judge the issues that affect
them. I think that WebTV showed this with the use of focus groups. More and
more user-interface companies are falling toward this model as well.

By the way, this isn't just about HTML. Java, JavaScript, ActiveX, etc. are
all in this boat. These pseudo-open standards languages aren't that open,
and each has its own set of issues which need to be addressed by their
developer community. But the developer needs a place in these discussions.

We are close, very close to being able to create true, real-time
applications that work cross-platform. If only we could gain a tad more
control over our medium, we as developers could do amazing things with the
technology in front of us.

--David Heller
dheller@tisny.com