Date: Mon, 01 Jan 2001 23:29:44 +0800
From: Walter Scott <[EMAIL PROTECTED]>
Subject: On Reconsideration Of Bobcat Vs. Net-Tamer


Previously, I wrote of "disappointment" regarding lack of sufficient 
memory to conveniently use Bobcat as an alternate web browser in place 
of Net-Tamer where Net-Tamer can't cut it on some websites. Nothing 
has changed as to my understanding of the "exhausted" memory issue 
between then and now. However, to be fair, I should indicate there are 
workarounds in Bobcat that allow it to do what it otherwise would not 
do.

While it's still true that Bobcat easily runs out of memory if you try 
to select and display a webpage from a link on the current page you're 
viewing -- that's to say you run out of memory when you view a webpage 
and then attempt to choose a link on that page to yet another page -- 
Bobcat's "external" program execution options allow a number of scenarios 
limited only by imagination. The author of Bobcat envisions and 
describes how to implement, for example, a means to view a webpage that 
Bobcat would not previously load due to exhausted memory by spawning a 
second or child session of Bobcat where the URL from the last attempt 
to load a webpage is extracted and called by the child session of 
Bobcat with recycled memory. It's a kind of Rube Goldberg process. 
But, it DOES work. In addition, Bobcat has a [D]ownload option that 
allows you to simply download the HTML content of a webpage to a file 
on disk without displaying that content. This has almost no impact 
memory usage. You can then look at those files later in Bobcat without 
the memory-sucking packet driver and terminal program running, or you 
can use a separate HTML viewer/converter outside of Bobcat. You could 
also call such viewer/converter software as an automatic external 
program. Implementation is not easy, though. It will require reading 
several documentation files and then writing plus debugging batch 
files which will ultimately do the job of executing an external task.

So, I want it to be clear that, in my last posting, the disappointment 
is not in the author of Bobcat and his efforts. It seems he has made 
it a priority to solve the memory issue I've addressed here. My 
disappointment is over the difficulty for writing a *GOOD* **DOS** web 
browser for minimalist/legacy hardware. There's too much code and too 
much functionality to cram into conventional memory. Thus, I 
understand even better David Colston's problems in supporting a web 
browser than I once did. Perhaps there's no way to have a reasonably 
decent text-based DOS web browser for minimalist/legacy hardware 
crammed into two executable modules that principally or only uses 
conventional memory. That's the point to consider here. Bobcat is a 
good web browser. But look at its size and how much memory it consumes 
to do all that it does. Look at how large Arachne is and its present 
requirement for both conventional and EMS/XMS memory in order to do 
all that it can do at a reasonable pace. Microsoft and IBM did us all 
some disservice over the many years they knew about "virtual memory" 
in other operating systems but refused to implement an equivalent for 
DOS. They came to the party, if you will, all too late.

When you think about it, there's possibly a single competitor to the 
functionality of Net-Tamer with service to those running legacy 
hardware. But that's not the primary goal of its creators -- E. G. the 
software is made to run on an 80186 CPU in handheld computers which 
means it can also run on 8086s 8088s and 80286s. That software -- a 
purely commercial package -- is split across several executable 
modules to achieve what Net-Tamer does in only TWO modules. Perhaps 
this is the route David Colston should have taken years ago. Ah, but 
hindsight is 20:20 and it may be too late to take advantage of it.

                        Walter Scott

<><><> News & Views, 24 Hours A Day, 7 Days A Week, Are The Net <><><>

NTReader v0.31(P)/Beta (Registered) in conjunction with Net-Tamer.



********************************************************
To unsubscribe from this list,
send a message to [EMAIL PROTECTED] with the single word
                     Unsubscribe
as the subject.
You MUST use the same address with which you subscribed!
********************************************************

Reply via email to