Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Computing > NZ Computing > Re: Car Recalls And Computer Geeks

Reply
Thread Tools

Re: Car Recalls And Computer Geeks

 
 
Bruce Simpson
Guest
Posts: n/a
 
      08-15-2003
On Fri, 15 Aug 2003 10:15:35 GMT, David Pears
<(E-Mail Removed)> wrote:

>On Fri, 15 Aug 2003 21:20:33 +1200, Bruce Simpson
><(E-Mail Removed)> wrote:
>
>>Okay, but what if you're on Jetstream 500 or some other "pay by volume
>>when over-cap" Net connection and the very regular (and sometimes
>>quite large) downloads are costing you $0.20 per MB. Why should you
>>have to *pay extra* just so that the product you've paid good money
>>for is brought up to scratch and in line with the promises made BEFORE
>>you bought it?

>
>So what do you suggest as an alternative? If you want someone to come
>to your house and do it for you without connecting to the net then
>that is possible, but you're going to pay a lot more for your
>software. How much extra are you willing to pay.


Let's see -- Microsoft makes more profit than Ford and GM combined.
Microsoft has accumulated cash reserves of over US50 billion.

I'd hardly say (based on these figrues) that Microsoft's prices are
already pared to the bone would you?

But I don't suggest we have Bill Gates come around at tea time and
install your patches. What I suggest is that Microsoft honour their
"Trustworthy Computing" promise.

Forget about adding new features or tarting up Media Player for the
time being and go back to basics:

1. Ensure that products ship with options turned OFF by default
instead of turned on. If RPC weren't a standard part of the XP
package then this problem wouldn't have happened -- and what home user
actually has any need for RPC on their home PC anyway????

2. Get an *ndependent*code audit done on *all* the existing code.
Buffer overrun errors are easy to check and trap.

And there are other things that could be done to improve the
"robustness" of their offerings but I charge for consulting

>And Linux is not an alternative. It has as many or more bugs that
>require patching. And they're not offering to reimburse your network
>costs either.


Hey, not true. You can get a 200% rebate on Linux for every bug you
find. Of course 200% of $0.00 is nothing -- but let's see Microsoft
equal that offer (in percentage terms

>Actually I'm the IT security manager for a population of around 12,000
>people. I've done little else over the past three days other than
>coordinate our response to Blaster. So I'm looking at the big picture
>as well, but don't see an easy fix. There is always the keep the
>mainframe argument, with its conservative and stable software, but
>people aren't willing to keep paying for it. So what's your quick and
>easy solution that doesn't cost us any more than we pay at the moment?


Exactly the same as 1 and 2 above. Does your network make extensive
(or any) use of RPC on the workstations? I expect not -- so why are
you installing it?? Because Microsoft are daft enough to include it
as a core component of WinXP??

Or have I got this all wrong?

--
you can contact me via http://aardvark.co.nz/contact/
 
Reply With Quote
 
 
 
 
David Pears
Guest
Posts: n/a
 
      08-15-2003
On Fri, 15 Aug 2003 22:24:17 +1200, Bruce Simpson
<(E-Mail Removed)> wrote:

>>>Okay, but what if you're on Jetstream 500 or some other "pay by volume
>>>when over-cap" Net connection and the very regular (and sometimes
>>>quite large) downloads are costing you $0.20 per MB. Why should you
>>>have to *pay extra* just so that the product you've paid good money
>>>for is brought up to scratch and in line with the promises made BEFORE
>>>you bought it?

>>
>>So what do you suggest as an alternative? If you want someone to come
>>to your house and do it for you without connecting to the net then
>>that is possible, but you're going to pay a lot more for your
>>software. How much extra are you willing to pay.

>
>Let's see -- Microsoft makes more profit than Ford and GM combined.
>Microsoft has accumulated cash reserves of over US50 billion.
>
>I'd hardly say (based on these figrues) that Microsoft's prices are
>already pared to the bone would you?


Microsoft's prices are what users are prepared to pay, even tho there
is a free alternative available. I presume the users think they're
receiving value for money, or they'd run something else. They seem
(mostly) to not want to run the more stable, secure, and expensive
mainframe operating systems, even tho you can pop an emulator card in
a PC and run them.

>But I don't suggest we have Bill Gates come around at tea time and
>install your patches. What I suggest is that Microsoft honour their
>"Trustworthy Computing" promise.
>
>Forget about adding new features or tarting up Media Player for the
>time being and go back to basics:


I suspect that people want a tarted up Media Player. And that if
Microsoft didn't supply one, then the competition would.

>1. Ensure that products ship with options turned OFF by default
>instead of turned on. If RPC weren't a standard part of the XP
>package then this problem wouldn't have happened -- and what home user
>actually has any need for RPC on their home PC anyway????


The problem is a little wider than just a single feature. How many
features do you disable before users start complaining that the
functions they want to use are difficult to install and find? And how
far do you go to protect users from (potentially) harming themselves,
like voluntarily installing a peer to peer file sharing program, or
something with spyware?

>2. Get an *ndependent*code audit done on *all* the existing code.
>Buffer overrun errors are easy to check and trap.


They might well be easy to check and trap. So why are there so many in
both Windows and Linux? And why should these independent programmers
be better at spotting them than the guys working for Microsoft and
developing Linux? And what about all the vulnerabilities that have
nothing to do with buffer overruns?

>And there are other things that could be done to improve the
>"robustness" of their offerings but I charge for consulting


And Microsoft has decided not to take up your offer. Whereas it does
pay hundreds (thousands?) of programmers already. The market has
decided your worth. (Nothing personal Bruce, as they're not giving me
millions of bucks worth of stock options either. But it is easier to
point out errors that are known and already patched, rather than to go
thru hundreds of thousands of lines of code looking for them.)

>>And Linux is not an alternative. It has as many or more bugs that
>>require patching. And they're not offering to reimburse your network
>>costs either.

>
>Hey, not true. You can get a 200% rebate on Linux for every bug you
>find. Of course 200% of $0.00 is nothing -- but let's see Microsoft
>equal that offer (in percentage terms
>
>>Actually I'm the IT security manager for a population of around 12,000
>>people. I've done little else over the past three days other than
>>coordinate our response to Blaster. So I'm looking at the big picture
>>as well, but don't see an easy fix. There is always the keep the
>>mainframe argument, with its conservative and stable software, but
>>people aren't willing to keep paying for it. So what's your quick and
>>easy solution that doesn't cost us any more than we pay at the moment?

>
>Exactly the same as 1 and 2 above. Does your network make extensive
>(or any) use of RPC on the workstations? I expect not -- so why are
>you installing it?? Because Microsoft are daft enough to include it
>as a core component of WinXP??


We have a core SOE which is pretty conservative and robust. But last
time I asked for a scan of all systems and a list of what software we
had installed, it amounted to about 50 pages of printout, 1 line per
program. Were any of them using RPC? Buggered if I know.

Now you could argue that we should restrict the software we let people
run and validate the stuff before we let people install it. But we
don't see ourselves as IT policeman there to stop people using tools
they feel help them do their jobs. We don't have the resources to
evaluate and integrate 2,500 different software packages. And we don't
even object to games, of which there were several hundred different
varieties on the printout... we have nurses and teachers working out
in remote communities where the nearest cinema is 1000 kilometers
away, and if a few computer games keep them on the job a few extra
months, then we're all in favour.

>Or have I got this all wrong?


I think you underestimate the complexity of an environment with
thousands of different users, all doing different things. Let alone a
user population of hundreds of millions of users, doing really wierd
****. Which is what Microsoft has to do.

David
 
Reply With Quote
 
 
 
 
E. Scrooge
Guest
Posts: n/a
 
      08-15-2003

"Bruce Simpson" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
> On Thu, 14 Aug 2003 22:03:51 +1200, "E. Scrooge"
> <(E-Mail Removed) (remove eye)> wrote:
>
> >> One would expect that parts AND LABOUR would be covered in the case of
> >> a defective product.

> >
> >It would only seem reasonable that all cars should be fitted with radar
> >sensors to help prevent any contacts when trying to park the car. A rear
> >view video could be handy as well. There are a number of options that

cars
> >could be fitted with but simply aren't.

>
> [big snip]
>
> True, there are many optional features that, if they're not fitted,
> you can't complain about.
>
> However, let me quote you from the Microsoft Website in respect to
> Windows XP:
>
> "Small business owners will be able to securely connect their Windows
> XP Professional computers to the Internet"
>
> Now it seems to me that this is being sold to customers as a standard
> feature of the product.
>
> So, when you find that you can't securely connect your Windows XP
> Professional product to the Internet, doesn't that mean it's faulty?
>
> Since computer software doesn't "wear out" then it must have been
> faulty from day one -- which means a design fault. It must therefore
> be the responsibility of the manufacturer to put right the fault at no
> cost to the user (including labour).
>
> That's all I'd expect Microsoft to do -- fix the system so that it
> meets the claims made for it, but without me (or a thousand other
> system admins) having to spend their own valuable (chargeable) time to
> do so.
>
> >On a side note, Bruce. I was looking at your most impressive Jet cart
> >videos . Do you have any idea just how hot your Jet engine gets? By
> >the look of the videos it has quite a hot looking glow to it. With your
> >foot on the brake it hardly goes quick enough to help keep the engine a

bit
> >cooler. Doing a 100 mph might help drop the temp a couple of degrees -
> >wouldn't do much for the nerves though.

>
> The engine runs at about 800-900 deg C. And don't worry, I'll have a
> *much* faster pulse-jet powered vehicle due for public display early
> next year. Richard Noble has endorsed the new project and may come
> out to see it run.


Just what the bros need to cook their weenies with.

E. Scrooge


 
Reply With Quote
 
Howard Johnson
Guest
Posts: n/a
 
      08-15-2003

"David Pears" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed)...
>
> So what do you suggest as an alternative? If you want someone to come
> to your house and do it for you without connecting to the net then
> that is possible, but you're going to pay a lot more for your
> software. How much extra are you willing to pay.


A lot of people without the skills will probably have to do that.
>
> And Linux is not an alternative. It has as many or more bugs that
> require patching. And they're not offering to reimburse your network
> costs either.


It does indicate the necessity for some distro of Gnu/Linux or Macintosh OSX
or BSD to _become_ a suitable alternative.
It is technically possible.



 
Reply With Quote
 
Max Burke
Guest
Posts: n/a
 
      08-15-2003
> Bruce Simpson scribbled:

>> Max Burke wrote:


> I consider this to be a bit like the spam argument. "just patch the
> bugs" is a direct equivalent to "just press delete".
> Individual patches and individual spam messages might be trivial and
> of little consequence.
> However, once they start piling up and becoming a regular irritation
> and expense then it's only natural that one's perspective might
> change.


>> Do you feel the same way and same 'perspective' about OSS/Linux
>> Bruce? Because OSS/Linux has just as many if not more bugs that need
>> weekly if not daily patching according to several Linux websites I
>> read regularly, and several email newsletters I get... And heres
>> the kicker; They're often the same type of bugs and resultant
>> patches that happen in Windows software....


> There's one very significant difference between OSS and Windows -- the
> amount of money you pay to acquire it.


I didn't pay anything for the Microsoft OS running on my computer. It
came with the system......

> OSS is free so you can always get a full refund at any time.


Yeah a refund of $00.00c sure is 'adequate compensation' for all the
time it takes and problems users *can* have when trying to get *nix/OSS
to do what it's supposed to do.....
ROTFLOL

> Windows
> is expensive and therefore those who hand over their cash ought to
> have the right to expect that it will work as advertised and, if it
> doesn't, that it will be fixed at no cost to them (either in the time
> taken to download/install patches or the cost of bandwidth to do so).


I have never had to pay Microsoft to download and install the *free
patches/updates* for their various OS'es.
Are you saying you have been paying Microsoft to do that? Why?
As for my 'bandwidth costs' I pay $19.95 a month and that covers ALL my
usage of bandwidth (for doing things like reading Aardvark each day),
NOT just downloading and applying updates to the OS.

> The root of my entire argument is that every other product you pay
> good money for comes with a level of support that says "if it doesn't
> work as claimed, the seller has a responsibility to put it right at no
> cost to the customer" -- so why should software be exempt??


Is it exempt? Not when I see Microsoft providing a *FREE SERVICE* to
download and install updates to fix problems with their products.....
Here's where you can find and use this FREE service Bruce:
http://v4.windowsupdate.microsoft.com/en/default.asp
http://www.microsoft.com/technet/security/default.asp
On top of that they provide free email newsletters and advisories to
*ANY ONE* who wants to subscribe.....


--
mlvburke@#%&*.net.nz
Replace the obvious with paradise to email me.
See Found Images at:
http://homepages.paradise.net.nz/~mlvburke

 
Reply With Quote
 
Max Burke
Guest
Posts: n/a
 
      08-15-2003
> Howard Johnson scribbled:

>> "David Pears" <(E-Mail Removed)> wrote in message
>> news:(E-Mail Removed)...


>> So what do you suggest as an alternative? If you want someone to
>> come to your house and do it for you without connecting to the net
>> then that is possible, but you're going to pay a lot more for your
>> software. How much extra are you willing to pay.


> A lot of people without the skills will probably have to do that.


>> And Linux is not an alternative. It has as many or more bugs that
>> require patching. And they're not offering to reimburse your network
>> costs either.


> It does indicate the necessity for some distro of Gnu/Linux or
> Macintosh OSX or BSD to _become_ a suitable alternative.
> It is technically possible.


How so? If the 'alternative' has the same bugs and problems as the OS
it's supposed to replace, then it's not really an alternative that
removes the problem is it....
And before you say OSS/*nix doesn't have those buffer overruns,
compromises, and bugs like Windows does I *can* provide the links
*again* that clearly shows the OSS/*nix alternative being offered has a
record of bugs that is just as bad as many like to claim windows
is.......

But on top of that OSS/*nix requires the user to do MAJOR
reconfigurations of the OS and applications to get them to work the way
they're supposed to work most of the time. The ordinary computer user
doesn't want that neither should they have to do that to use a
computer.....

--
mlvburke@#%&*.net.nz
Replace the obvious with paradise to email me.
See Found Images at:
http://homepages.paradise.net.nz/~mlvburke

 
Reply With Quote
 
Nicholas Sherlock
Guest
Posts: n/a
 
      08-15-2003
Bruce Simpson wrote:
> Likewise -- what are the costs if you're the admin of a network with
> anywhere from five to a thousand or more PCs, ensuring that all the
> boxes under your control are always patched to the latest release


Exactly the same in bandwidth terms. Install Windows Update Server on the
main machine, and you'll only have to download the patches once.

Cheers,
Nicholas Sherlock


 
Reply With Quote
 
Bruce Simpson
Guest
Posts: n/a
 
      08-16-2003
On Sat, 16 Aug 2003 11:27:42 +1200, "Nicholas Sherlock"
<(E-Mail Removed)> wrote:

>Bruce Simpson wrote:
>> Likewise -- what are the costs if you're the admin of a network with
>> anywhere from five to a thousand or more PCs, ensuring that all the
>> boxes under your control are always patched to the latest release

>
>Exactly the same in bandwidth terms. Install Windows Update Server on the
>main machine, and you'll only have to download the patches once.


But we must asume that the admin gets paid a salary ($$) and that a
*sensible* admin will quarantine and fully test updates to ensure that
there aren't any adverse side-effects (not unknown with updates from
MS, particularly those focused on patching holes).

All of that takes time (and therefore costs the company who bought the
software) signifciant money.

If they bought a fleet of Holden Commodores for their executives and
salespeople then Holden found a major design/manufacturing fault,
would the company be expected to pay its own mechanic to fix the
problem?

I think not.

--
you can contact me via http://aardvark.co.nz/contact/
 
Reply With Quote
 
Enkidu
Guest
Posts: n/a
 
      08-16-2003
On Fri, 15 Aug 2003 10:15:35 GMT, David Pears
<(E-Mail Removed)> wrote:
>
> There is always the keep the mainframe argument, with its
> conservative and stable software, but people aren't willing to
> keep paying for it.
>

LOL! I was the Systems Programmer on several large MVS setups. Patch
tapes came in every week to be installed, and we paid IBM through the
nose for the service! In the early days of MVS it made Windows version
1 look rock stable. The only clues were normally buried in a two foot
stack of striped paper. Very often a patch would stop the machine
IPLing, or it would just crash the machine after a few minutes.

I would strongly disagree that mainframes have been traditionally more
stable than current Windows Server platforms, from NT on. It always
amazes me that people say so, and I've been at the coalface in both
environments.

Where IBM had it sussed was in the bug reporting and fix distribution
areas. Every bug had its APAR and every fix had its PTF. Percolation
of error handling allowed crippled systems to stay up longer. But
receiving, installing and moving PUT tape levels into production was a
full-time job.

IBM said, I believe that the first MVS had 2 million lines of code,
and exactly the same number of bugs. I think we do better these days
with many more lines of code and less bugs.

Cheers,

Cliff
--

Signed and sealed with Great Seal of the Executive
Council of the Internet, by The Master of The Net.
 
Reply With Quote
 
David Pears
Guest
Posts: n/a
 
      08-16-2003
On Sat, 16 Aug 2003 14:20:04 +1200, Enkidu <(E-Mail Removed)> wrote:

>> There is always the keep the mainframe argument, with its
>> conservative and stable software, but people aren't willing to
>> keep paying for it.
>>

>LOL! I was the Systems Programmer on several large MVS setups. Patch
>tapes came in every week to be installed, and we paid IBM through the
>nose for the service! In the early days of MVS it made Windows version
>1 look rock stable. The only clues were normally buried in a two foot
>stack of striped paper. Very often a patch would stop the machine
>IPLing, or it would just crash the machine after a few minutes.
>
>I would strongly disagree that mainframes have been traditionally more
>stable than current Windows Server platforms, from NT on. It always
>amazes me that people say so, and I've been at the coalface in both
>environments.
>
>Where IBM had it sussed was in the bug reporting and fix distribution
>areas. Every bug had its APAR and every fix had its PTF. Percolation
>of error handling allowed crippled systems to stay up longer. But
>receiving, installing and moving PUT tape levels into production was a
>full-time job.
>
>IBM said, I believe that the first MVS had 2 million lines of code,
>and exactly the same number of bugs. I think we do better these days
>with many more lines of code and less bugs.


Things change. When I started as a VM sysprog, in 1985 or so, we had
an abnormal abend about once a month or so. By the time I was at the
European Space Agency in 1992, I ran a system for a whole year with
only one abend. And that was me thinking I could zap real memory and
get away with it... and I did, for about 5 seconds By the time I
left IBM Netherlands in 2000, the system I developed security software
for had gone five years without an unsheduled outage. And that was a
400,000 user system, and so was pretty complex.

Most of the improvement was in the software quality. But the
consolidation of small data centers into megaplexes, as they were
calling them, got rid of the cowboys. Such as myself. The sysprogs
left running the show were pretty professional and did a lot of
testing. Testing never really interested me.

David
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
CAR INSURANCE FOR YOUR CAR Ahmed Digital Photography 3 10-31-2007 10:50 AM
CAR INSURANCE FOR YOUR CAR Ahmed C++ 0 10-31-2007 02:25 AM
CAR INSURANCE FOR YOUR CAR Ahmed C Programming 0 10-31-2007 02:24 AM
Re: Car Recalls And Computer Geeks Gavin Tunney NZ Computing 4 08-14-2003 11:11 PM



Advertisments