Subject: Re: FWD: CompuServe's Web browser From: R Ballard Date: Tue, 2 May 1995 23:44:35 -0400 (EDT)
How the Web Was Won
Subject: Re: FWD: CompuServe's Web browser From: R Ballard Date: Tue, 2 May 1995 23:44:35 -0400 (EDT)
In-Reply-To: <9504211223.AA10548@ InterServ.Com>
Message-ID: 
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Status: O
X-Status: 



On Fri, 21 Apr 1995 gnewton@interserv.com wrote:

> The facts are that on installation CNL.exe (the CIS Web installation program) 
> searches your hard drive for winsock.dll - if it exists - and renames it 
> winsock.000 (although it also gives you the option to rename it anything else 
> you like).  It then writes its own winsock.dll to your /windows subdirectory. 
>  As an I-Box user I found that all my I-Box services were then directed 
> through Compuserve, which, as I said in another post yesterday, is sometimes 
> slower and sometimes faster.

> If you wish - as I did - to go back to using the I-Box services, it was a 
> simple matter of renaming the two files; i.e. the I-Box winsock became 
> winsock.dll again, and the C'Serve version as...well, you work it out.  
> Situation returned to normal. 

This is great if you are using I-Box, but if you are using NFS capable 
software such as Frontier, the results can be catastrophic.  I once spent 
$1000 trying to unravel two "gotta own it" packages that insisted on 
having their own flavor of winsock.dll, eventually I had to go back to 
Trumpet.

If the interface clobbered was the Microsoft TCP/IP VXD flavor, you would 
have taken out the entire operating system kernal (I did that too).

> Given the complexities of the matter this installation routine is far from 
> being the most obnoxious I've ever encountered; the winsock renaming was 
> clearly documented in the readme and other files, and it worked as advertised. 
Except that it shredded the other browsers, and could have rendered 
servers or ethernet strings in a state of chaos.

>  The documentation did not however, to the best of my recollection, clearly 
> document the impact of the change, and the 'fix'; however I'm no systems maven 
> and I managed to work it out for myself. To the conspiracy theorists I have 
> just one piece of advice: RTFM.  (Read the (fill in your own word) manual).
There are about 2 million users, many of whom are quite knowledgible, who 
are about to try and let CIM connect them to the internet and could 
seriously damage corporate networks.  The likelihood is that compuserve 
will not be blamed, the internet will.

> On Friday April 21 Don Taylor wrote:
> >I find it extremely hard to believe that anyone like CIS would be so
> stupidly arrogant to do something like that.(big snip...)I eagerly await the 
> facts.

I am excited to see that CIS is offering internet connectivity, but I am 
concerned that CIS may did not consider the impact of insisting on their 
own stack and browser.

	Rex Ballard

From rballard@cnj.digex.net Wed May  3 00:01:25 1995