| Thursday,
December 7, 2000 When
I first got this I couldn't figure out what it was about, but the search
engine showed me it was in the mail archives, to be exact mail 60. I had
written something about problems with Netscape, and used a dramatic
opening line. The actual header was supplied by the editors:
"NETSCAPE MUST DIE." When Mr. Osmont sent me his
complaint, which was certainly not marked as private, he included that
header. I posted his letter in Mail 60, and reprint it below. He has
since replied:
Sir,
I have found on yahoo a link to your web site with a
page containing my name and an email message I had sent to you a while ago
regarding a disturbing article that you had published in some magazine.
Obviously, I find once again your action very
disapointing, as you are not only publishing some communication that was
intended to be private, but you are also taking the message out of his
context by not publishing the one I was responding to. You actually make
it sound like I wrote this horrible title, but I was in fact just
responding to this very aggressive article of yours.
These is not the type of act that one would expect
from a colomnist or journalist. On the other hand, my email is not the
type of email a company would expect from one of his employees. I was
shocked to see this message, as I could not believe I wrote this. Probably
a moment of deep frustration expressed a little to loud.
Anyhow, I would appreciate that you include your
article on front of my message or remove this message from your web site.
Thank You.
Stephan Osmont
So, I wish it made clear that Mr. Osmont did not
supply the header. Here is his original letter. The NETSCAPE MUST DIE
title is NOT his. His original letter had a Netscape return address. This
latest one does not.
Netscape Must Die!
Your article is pathetic. Did it even come to
your mind to question NT's reliability? Shouldn't such a strong OS be able
to handle this problem?
Anyway, the point is that people at Netscape or
Microsoft deserve respect for what they have accomplished and for what
they brought to the world.
On the other hand, whiners like you must die.
Stephan Osmont
Do note that I have more than once spoken of
unreliabilities in NT, Windows, and much else.
You can find the original "NETSCAPE
MUST DIE" article at
http://www.byte.com/column/chaosmanor/BYT19990720S0001
It's pretty clear from reading the column that
the title was supplied by the editors, and it's also clear why they did
that. In any event I want it clearly understood that Mr. Osmont did not
write the title "NETSCAPE MUST DIE" in his letter sent to me and
published 18 months ago, and he has my apologies. As to his "private
mail" I think my mail policies are clearly stated. I don't recall if
he sent his commentary to me or to BYTE, but given the content it hardly
matters, and it certainly doesn't read as private correspondence. When you
send letters to authors or their editors commenting on published material,
you must expect them to believe you intend those to be published.
Incidentally, what had annoyed me about Netscape
(which I used rather than Internet Explorer in those days) was that I had
tried to upgrade it. The result was horrible. Among the things I said in
that column:
"What is happening is that Netscape's install program doesn't
check to see if it has sufficient resources to run before it launches.
Once launched, it grabs what it can and if that isn't enough -- it won't
be if you have many other processes going -- it blows up instead of
shutting down with a message saying close all the other windows.
"One of the things that can blow it up is Netscape running, but of
course if you used Netscape's "smart update", then Netscape will
be running."
There was more, and it's all there in the old
column. Since then I gather that Netscape has cleaned up its act a
bit.
Dear Jerry,
Regarding your past problems upgrading Netscape via
interactive online file transfers:
I also had problems when I tried to do an online
upgrade from (approximately) 4.0 to 4.5. I dumped the idea of an online
upgrade and ordered a CD from Netscape. This approach worked much better,
and I got good service from version 4.5. Compatibility was very good for
it, and I would resort to Netscape when Opera (my favorite) wasn't
compatible with the latest jazzy html somewhere.
Later I ordered a CD for Internet Explorer,
installed it in kitchen sink mode (to see what they had to offer), and all
was well--although I found myself pruning off some of the pieces of the
kitchen sink, just to keep life simple. :-) Internet Explorer 5.0 seems to
work well.
I have recently received a CD for Netscape 6 and
installed it. Again I asked for a kitchen sink installation. There were no
installation problems. I'm evaluating (and pruning) the new Netscape
browser now. It seems to work
Based on these experiences, I recommend the CD
approach for Microsoft and Netscape browser upgrades. Opera upgrades are
performed using simple file transfers across the network with locally
controlled installation processes. I have performed a number of Opera
upgrades with no problems.
Concerning the "Netscape Must Die!" title:
a little hyperbole does no harm here.
Robert Griswold
I agree completely. But if Netscape had just made
it easier to save their upgrade rather than insisting on running it,
things would have been much better. I no longer use Netscape. Am I missing
much?
And now this:
Sir:
I think the points made in the editorial are valid,
but applied to the wrong party. It is increasingly clear to anyone not a
partisan that the majority of votes, not just in the country as a whole,
but in Florida specifically, were cast for Al Gore. The republican
response to that has been to loudly ask that this all be put behind us,
for the good of the country, and just have finality.
I have heard the phrase "stealing votes"
used, repeatedly, by Republican loyalists, and I have this question: If a
person votes for a candidate, and we can, through manual recounts,
determine who they voted for, how is that "stealing" anything?
Yes, I am disturbed by the lack of an objective standard for manual
recounts. Shouldn't the answer to that be to impose an objective standard,
not to abandon the idea of an accurate count? Gov. Bush signed a law
specifically allowing "dimpled" chads to be counted in hand
recounts in Texas. It is the height of hypocrisy to now claim that such
ballots cannot be accurately counted.
No, I believe that Bush will be the next president.
I will obey his orders, and treat him with respect. I will also know that
he was not selected (I do not say elected here for a reason) by the people
of this country. I would take no issue (other than a philosophical one)
with the popular vote being different than the electoral vote. I do take
issue with the grandstanding, name calling, finger pointing being done by
the Republicans. Their actions on the absentee ballots was certainly
outside the law, if not specifically illegal, in Martin and Seminole
county. I do not think either county should invalidate any votes, despite
that being the specific statutory remedy. In those cases, there is no
doubt that the people had a right to vote, and the vote is accurate. No
technical violation should overturn those votes, no matter how offensive
the conduct was in getting those votes. The Republicans are apparently
interested in the outcome only, not whether it is the will of the voters,
or has any resemblence to accuracy.
I dread the day, and I believe it certainly
possible, in the near future when the DOJ discovers actual voter fraud by
the republican party or its agents. It will present a crisis for the Bush
presidency that no presidency should have to endure.
My conclusion, contrary to that of the editorial of
Mr. Praeger, is that the people will grow tired of the rhetoric of the
Republicans given their actions, and the elections in two years will be a
sweep toward the Democrats.
Please omit my email address and name, for obvious
reasons, if you publish this.
Well, I don't agree. It is true that Texas seems
to allow "dimpled chad" votes; no other jurisdiction does, and
certainly Florida never contemplated doing so. My insistence is that
whatever rules are used, they be agreed to in advance; and if there are to
be rule changes they be changes that at least MIGHT have been agreed to in
advance. California law says a "three corner punched" chad is a
punch, a two-corner is ambiguous, and a one-corner is NOT a vote. There is
no mention of dimples at all. California uses the same machinery as
Florida, and has had a Democratic legislature for some time.
I think the Republicans are insisting largely on
application of rules in place at the time of the election. As to
"fraud", of course there was fraud, on both sides, and all over
the country. There never was an election in which there wasn't. Usually it
doesn't matter, but sometimes, as in 1960, it does. But while we can hope
to keep partisan fraud down, it's going to happen. That's the nature of
the power game. Those who feel strongly tend to dominate parties, and that
won't change, because those who don't feel strongly don't compete for
party office. It is one of the oldest dilemmas of power: make any office
powerful enough and it will tend to be occupied by those unscrupulous
enough to do anything to get it; in the end, that can extend to assassinations,
locking people in dungeons, murder of siblings, parricide, and Herod's
Slaughter of the Innocents. Fortunately we haven't got there yet.
What I really wish would happen is that we would
repeal about a million pages of laws and regulations so that Washington
would become a sleepy Southern village that commands the Armed Services
and conducts foreign policy and otherwise leaves running the country to
the states. Of course that won't happen. But I grew up in a time when it
was far more relevant to my life who was County Supervisor, or Sheriff,
than who was President of the United States; which, except in war time, is
the way it ought to be. In my not very humble opinion.
I just finished an article in Salon that compares
the current state of the software/hardware industry to that of the US car
industry of the 50's and 60's. Simply stated, there are ways to design
software so that there are several orders of magnitude fewer defects
(bugs) than currently are shipped (and evidently cheaper). The prognosis,
only foreign software will cause any change in US software design.
On another topic, you had said you would be getting
an Apple G4. If you haven't done so yet, they should be very cheap once
Apple puts in place its inventory clearing sales/rebates.
Intellectual Capital style columns most definitely
would cause me to subscribe, even if I disagree with a lot of what you say
in the articles.
Don Scherer
Is there likely to be any foreign software? I
certainly would not object to having competition. Microsoft needs that
badly.
As to software design, I think the real culprit
is C. C is more assembler than higher level language, and now that we have
new hardware with great potential, I think a more structured and readable
language, such as Modula-2 would improve our software enormously by making
it a lot clearer what it is DOING. I really ought to start the language
debates again. I never thought C was the proper language for writing complex
programs.
I don't entirely disagree with the notion that C and
it's variants are to blame for poor code. Any compiled language that
doesn't (by default anyway) complain when you toss one data type into
another by making some grand assumption that you know what you are doing
is bound to be a loaded gun.
On the Gripping Hand, however, I believe that
a good portion of the problem is lack of education in the Computer
Sciences. I feel (no exact proof mind you, just hiring experiences) that
most CS or related programs let good form and systematic design take a
back seat to having the latest software programs and whizbang products in
use in the curriculum so it can end up on a resume. I would take a person
schooled in Assembler or any other unstructured dialect if they would only
demonstrate some sense in design and systematic thought. The tools will
always change, the lessons learned are the real value.
My 2 cents anyway. Keep up the good work!
Marlin Roberts Clearwater Research
Thanks. I think it may be time to open the
language debates again. C will compile nonsense. Pascal and its
derivatives will not, and do range checking; it's harder to get a Delphi
or Pascal or Modula-2 program to compile, but when it does, it will
generally do what you expected it to do. With C you have to simulate the
compiler in your head, and for many that's a formula for making your head
explode...
I think it is time for readability.
TOP
|
CLICK HERE for
a Special Request. 
