 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> We had premptive multitasking operating systems and large C compilers and
> ray tracers and modellers and music sequencing software and complex
> computer games and so on and so forth.
If you're referring to your Amiga, then it was the same with my Acorn. But
those machines were orders of magnitude less complex than a PC today.
> I don't really see how the software M$ writes is any "bigger" or "more
> complex" than what existed before.
Ha. Compare what Windows XP does with your Amiga OS. Think about hardware
support, multi-processor support, multi-user support, multi-APIs for
everything, wireless networking, and a million other things.
> I've also seen software that's much more reliable. *cough* POV-Ray. When
> was the last time you saw it crash?
Apart from the beta, haven't seen it crash - but often it shows weird
results or fails to parse. And if such a relatively small program such as
POV with a tiny user-base has bugs, then I think MS has done pretty well
with Windows.
> Similarly, have you *ever* seen Linux crash? [A huge number of Linux
> applications are hopelessly buggy, but the OS itself seems rock-steady as
> far as I can tell.]
When was the last time you saw XP itself crash? I know I can't remember the
last time (and I use it regularly on 4 different machines with very
different hardware), apart from when there was faulty hardware.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> If M$ made a quality product and charged a lot of money for it, I wouldn't
> have a problem with that.
The MS shareholders would though, because they wouldn't make as much profit.
It's the same as if Ford suddenly decided all cars were going to be made to
Rolls-Royce standards - not exactly a smart business move.
> What I have a problem with is the fact that they charge a fortune for very
> low-quality products, and get away with it.
Last time I checked, a company was free to sell their products for whatever
price they wanted to. If they try to sell them too high, nobody will buy
them, if they try to sell them too low, they won't make any money. There's
a point where you make maximum profit - I assume MS are pretty close to that
point with their products.
> And they get away with it precisely because of the underhanded techniques
> they use to eliminate all competition.
Never seen any real competition for Windows, Linux is certainly getting a
lot better, if they could get people to release games for it (like they do
for xbox/ps3/pc) then they'd be a strong competitor I think. Not used
Apples much, but again lack of games prevents me from looking seriously at
it.
> Apparently, if you make enough money, the law does not apply to you...
No, it just means you can afford to pay the fines without going bankrupt.
Which in turn means you can take more risks in the first place.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott wrote:
> If you're referring to your Amiga, then it was the same with my Acorn.
> But those machines were orders of magnitude less complex than a PC today.
So what you're saying is that if hardware is complicated, it is
impossible to develop quality software?
>> I don't really see how the software M$ writes is any "bigger" or "more
>> complex" than what existed before.
>
> Ha. Compare what Windows XP does with your Amiga OS. Think about
> hardware support, multi-processor support, multi-user support,
> multi-APIs for everything, wireless networking, and a million other things.
Why do we *need* multi-APIs for everything in the first place?
Leaving aside hardware [bad drivers can screw up any OS], what does
Windoze actually do that AmigaDOS doesn't? Well, let's see now. It has
networking. It's multi-user and has access permissions. It... uh... no,
I'm struggling to think of anything else new it has. That seems to be
able it, really. (Unless you count IE as part of the OS.)
>> I've also seen software that's much more reliable. *cough* POV-Ray.
>> When was the last time you saw it crash?
>
> Apart from the beta, haven't seen it crash - but often it shows weird
> results or fails to parse. And if such a relatively small program such
> as POV with a tiny user-base has bugs, then I think MS has done pretty
> well with Windows.
I've seen it crash. I discovered that in some version or other, calling
exp() with a suitably large argument causes POV-Ray to instantly shut
down. This has since been fixed.
Almost all software possesses something you could describe as a "bug".
The point is, some bugs are more serious than others. Where I work, Word
is constantly crashing. The hours we've wasted because Word has crashed
and eaten somebody's work... I don't hear any stories of POV-Ray
crashing half way through a million-hour render.
>> Similarly, have you *ever* seen Linux crash? [A huge number of Linux
>> applications are hopelessly buggy, but the OS itself seems rock-steady
>> as far as I can tell.]
>
> When was the last time you saw XP itself crash? I know I can't remember
> the last time (and I use it regularly on 4 different machines with very
> different hardware), apart from when there was faulty hardware.
Interesting. When I got my laptop, it crashed within 14 *seconds* of
being turned on.
I will say this: As time has gone on, that laptop has crashed less and
less. So M$ are at least making some kind of positive difference.
But even so, here at work I look after a large cluster of PCs. And
usually in any given week at least one of them will give be a BSOD.
Indeed, just this weekend, one of the servers rebooted itself due to one...
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott wrote:
>> If M$ made a quality product and charged a lot of money for it, I
>> wouldn't have a problem with that.
>
> The MS shareholders would though, because they wouldn't make as much
> profit. It's the same as if Ford suddenly decided all cars were going to
> be made to Rolls-Royce standards - not exactly a smart business move.
No - but selling cars which you know are going to crash at least once
every 52 days... well, if a company tried to do that, they'd be shut
down. However, for software it seems it's OK to sell a malfunctioning
product.
>> What I have a problem with is the fact that they charge a fortune for
>> very low-quality products, and get away with it.
>
> Last time I checked, a company was free to sell their products for
> whatever price they wanted to. If they try to sell them too high,
> nobody will buy them.
Well, sure, if there were an alternative, people would run out and buy
that in their droves. I'm sure M$ would radically rethink their strategy
if that happened. But it won't.
>> And they get away with it precisely because of the underhanded
>> techniques they use to eliminate all competition.
>
> Never seen any real competition for Windows.
Perhaps not. (Does NetWare count? I don't really know much about it.)
But what about, say, Word? There's quite a few other word processors out
there - and in past times there were even more. And most of them were a
lot more reliable than Word...
> Linux is certainly getting
> a lot better, if they could get people to release games for it (like
> they do for xbox/ps3/pc) then they'd be a strong competitor I think.
> Not used Apples much, but again lack of games prevents me from looking
> seriously at it.
Apple requires you to buy new hardware, so it's not purely a software
decision. Linux is nice, but... well, it's fundamentally designed for
UNIX nerds. So if you're not, good luck... Besides, I'm not convinced
that the whole UNIX design is particularly coherant. (E.g., autoconf
exists.)
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> So what you're saying is that if hardware is complicated, it is impossible
> to develop quality software?
No, but with complicated hardware, users expect complicated software, and
with that comes bugs. Sure, even I could probably write a word processor
that rivalled what I had on my Acorn, and given a suitable amount of time
could make it pretty stable. But everyone would laugh at it.
> Why do we *need* multi-APIs for everything in the first place?
Dunno, just the way it is, because of the wide variety of hardware I guess.
At least for graphics there's only DirectX and OpenGL now, there used to be
3Dfx/Glide and some radeon thing IIRC. I don't know what alternatives there
are for DirectX for game controller input, sound, networking etc, but I'm
sure there are some in use.
> Leaving aside hardware
But I think that's one of the key points. Windows must work with every
possible piece of hardware that's available now, PLUS it has to work with
anything that might come out in the future... Can't imagine Amiga coming
out with a new 64-bit dual core machine, and then you being able to use the
same Amiga OS you had before with it...
> [bad drivers can screw up any OS], what does Windoze actually do that
> AmigaDOS doesn't? Well, let's see now. It has networking. It's multi-user
> and has access permissions. It... uh... no, I'm struggling to think of
> anything else new it has. That seems to be able it, really. (Unless you
> count IE as part of the OS.)
Just have a browse through the services running on your machine...
> Almost all software possesses something you could describe as a "bug". The
> point is, some bugs are more serious than others. Where I work, Word is
> constantly crashing.
Oh yeh, you're still running Word 1985 or whatever :-) Word only crashed
here when we had a buggy printer driver (but then so did any other
application that tried to print - it's just we mostly used Word).
> The hours we've wasted because Word has crashed and eaten somebody's
> work... I don't hear any stories of POV-Ray crashing half way through a
> million-hour render.
POV crashed on me during a long animation render, gave me a "variable not
declared" or something error, even though it'd just rendered 1000 copies of
the exact same SDL file previously...
> Interesting. When I got my laptop, it crashed within 14 *seconds* of being
> turned on.
And how many other XP machines exhibited this behaviour? I've used XP
(along with friends) right back from the pre-release versions of XP, and the
only time I've seen that behaviour is due to bad hardware. Once it was a
faulty RAM stick, another time was a dodgy hard drive, and the third time
was heat sink detachment!
> But even so, here at work I look after a large cluster of PCs. And usually
> in any given week at least one of them will give be a BSOD.
So what's the most common cause of BSODs then? Once a week on various
machines sounds like some broken software/driver that everyone has
installed...
> Indeed, just this weekend, one of the servers rebooted itself due to
> one...
And what was the reason for that? If it was my server I would want to find
out exactly which driver/program/hardware caused the BSOD.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Apple requires you to buy new hardware, so it's not purely a software
> decision. Linux is nice, but... well, it's fundamentally designed for UNIX
> nerds. So if you're not, good luck... Besides, I'm not convinced that the
> whole UNIX design is particularly coherant. (E.g., autoconf exists.)
It's just because nobody feels that could write something that is superior
enough to Windows to make enough money to warrant the development?
I mean, come on, Windows is pretty flexible, given that my mum can do stuff
on it, I can use it to run CFD simulations and design LCD screens, also I
can play 3D games and even write 3D games on it, plus raytrace :-)
ANd you even said yourself that XP is reliable now (although I always found
it reliable) - if you upgrade your ancient copy of Word you'll find that is
reliable too. Where exactly is any competition meant to improve to get
people to swap?
Sure, you can reach for niche markets, but you're never going to take over
from Windows with that approach.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott wrote:
> No, but with complicated hardware, users expect complicated software,
> and with that comes bugs. Sure, even I could probably write a word
> processor that rivalled what I had on my Acorn, and given a suitable
> amount of time could make it pretty stable. But everyone would laugh at
> it.
Personally, I don't think so. People have created products like
OpenOffice and KOffice and so forth, and they work. Reliably.
>> Leaving aside hardware
>
> But I think that's one of the key points. Windows must work with every
> possible piece of hardware that's available now, PLUS it has to work
> with anything that might come out in the future... Can't imagine Amiga
> coming out with a new 64-bit dual core machine, and then you being able
> to use the same Amiga OS you had before with it...
Interesing you should say that. You're aware of course that the fist
Amiga was a 16-bit machine, and the later ones were 32-bit? Or that
originally the graphics hardware only supported 12-bit colour, and later
added 24-bit colour? Or that originally you could only have 6 bits per
pixel and later you could have 8?
And - most importantly - you can take an application written back when
we only had 6 bits per pixel, and run it on a screen operating at 8 bits
per pixel. And it can share the screen with other applications and
change parts of the colour table and so on and so forth, and WORK.
[Assuming it does all this through the OS. The Big Problem the Amiga had
is that since the hardware is "always" the same, a lot of software
bypasses the OS. Obviously this breaks horribly when the hardware
changes...]
>> What does Windoze actually do that
>> AmigaDOS doesn't? Well, let's see now. It has networking. It's
>> multi-user and has access permissions. It... uh... no, I'm struggling
>> to think of anything else new it has. That seems to be able it,
>> really. (Unless you count IE as part of the OS.)
>
> Just have a browse through the services running on your machine...
So Windows is designed to do a bunch of unecessary stuff in the
background by default. I consider this a design flaw. What *useful*
stuff does Windows do that AmigaDOS doesn't?
>> Almost all software possesses something you could describe as a "bug".
>> The point is, some bugs are more serious than others. Where I work,
>> Word is constantly crashing.
>
> Oh yeh, you're still running Word 1985 or whatever :-) Word only
> crashed here when we had a buggy printer driver (but then so did any
> other application that tried to print - it's just we mostly used Word).
Word 2003, currently. Which, in fairness, seems slightly less buggy. But
still nowhere near as reliable as we'd like.
>> The hours we've wasted because Word has crashed and eaten somebody's
>> work... I don't hear any stories of POV-Ray crashing half way through
>> a million-hour render.
>
> POV crashed on me during a long animation render, gave me a "variable
> not declared" or something error, even though it'd just rendered 1000
> copies of the exact same SDL file previously...
That's not a crash, that's a parse error. Most likely due to a subtle
interaction in your SDL code...
>> Interesting. When I got my laptop, it crashed within 14 *seconds* of
>> being turned on.
>
> And how many other XP machines exhibited this behaviour? I've used XP
> (along with friends) right back from the pre-release versions of XP, and
> the only time I've seen that behaviour is due to bad hardware. Once it
> was a faulty RAM stick, another time was a dodgy hard drive, and the
> third time was heat sink detachment!
Don't forget MS Blaster. [The only virus I've ever had the misfortune to
come across. Nice how McAfee automatically removed it. Or at least, it
*would* have been nice if it had...]
>> But even so, here at work I look after a large cluster of PCs. And
>> usually in any given week at least one of them will give be a BSOD.
>
> So what's the most common cause of BSODs then?
And how in the name of God do you find out what *causes* one?
I mean, sure, it tells you what was happening at the exact instant the
crash happened, but how do you find out what the actual problem is?
All the STOP messages I get are either UNHANDLED_KMODE_EXCEPTION or
IRQL_NOT_LESS_OR_EQUAL. Could be anything.
> Once a week on various
> machines sounds like some broken software/driver that everyone has
> installed...
Yeah, probably. But how the hell do you find out what it is?
>> Indeed, just this weekend, one of the servers rebooted itself due to
>> one...
>
> And what was the reason for that? If it was my server I would want to
> find out exactly which driver/program/hardware caused the BSOD.
If there were anything I could do to find out, I'd do it. Unfortunately,
there isn't.
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
scott wrote:
> It's just because nobody feels that could write something that is
> superior enough to Windows to make enough money to warrant the development?
Indeed - primarily because if they somehow managed it, they would
instantly be bought out. The End.
> I mean, come on, Windows is pretty flexible, given that my mum can do
> stuff on it, I can use it to run CFD simulations and design LCD screens,
> also I can play 3D games and even write 3D games on it, plus raytrace :-)
Apart from your mum being able to work it, the same could be said for
just about any OS. (And even Linux isn't that hard to work these days.
Damn hard to set up, but not that hard to operate once you eventually
get it working.)
> ANd you even said yourself that XP is reliable now (although I always
> found it reliable) - if you upgrade your ancient copy of Word you'll
> find that is reliable too. Where exactly is any competition meant to
> improve to get people to swap?
True story: One of our users was working on a vital report. She hit
"Save", and Word crashed. From that point onwards, any attempt to access
this file caused Word to instantly crash. So I loaded it up in
OpenOffice, saved it again, and several hours of potentially lost work
were saved - literally.
Why would anybody use Word? 2 reasons:
1. Everybody has heard of Word. Nobody knows what OpenOffice is.
2. OpenOffice still has some rough edges to be smoothed out.
Any idea how many people are switching away from IE? Enough to provoke
M$ to start development work on it again.
Looks like competition is finding plenty to improve on...
> Sure, you can reach for niche markets, but you're never going to take
> over from Windows with that approach.
Sadly, it seems we are stuck with M$ Windows forever. I mean, M$ is too
powerful to beat. They're unstoppable. We're all doomed...
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> http://thedailywtf.com/Articles/SUCSESS.aspx
>
> So, so wrong...
>
> Of course, the *real* WTF is that final sentence: Even though this is
> complete garbage, I couldn't convince my boss because the application
> "works". (!!)
>
> If "programmers" like this were to ever get hold of Haskell... my God,
> it's too horrifying to think about! >_<
>
(setq t nil)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Personally, I don't think so. People have created products like OpenOffice
> and KOffice and so forth, and they work. Reliably.
For me MS Office works reliably - and it's not like I only use 5% of its
features.
> Interesing you should say that. You're aware of course that the fist Amiga
> was a 16-bit machine, and the later ones were 32-bit? Or that originally
> the graphics hardware only supported 12-bit colour, and later added 24-bit
> colour? Or that originally you could only have 6 bits per pixel and later
> you could have 8?
And you could use the same OS version with all these hardware
configurations?
>> Just have a browse through the services running on your machine...
>
> So Windows is designed to do a bunch of unecessary stuff in the background
> by default. I consider this a design flaw. What *useful* stuff does
> Windows do that AmigaDOS doesn't?
Virtual memory, automatic updates, protecting itself from nasty programs
that try to access things they shouldn't, providing common dialogs for all
applications, plug and play, CD/DVD-RW access, wireless networking, remote
desktop, configuring machines remotely, network file systems, network
printing, offline files, hibernating, file type associations, encryption,
working in multiple time zones, working with different region settings,
firewall. I'm sure there are more, but those are the most useful.
> That's not a crash, that's a parse error. Most likely due to a subtle
> interaction in your SDL code...
Nope, nothing was changing - the next time it stopped on a different frame.
I narrowed it down to some buggy handling of spline objects and arrays.
Just had to re-start the rendering on the exact frame whenever it stopped.
> And how in the name of God do you find out what *causes* one?
Gulp. Look in the event log. Use Google.
> I mean, sure, it tells you what was happening at the exact instant the
> crash happened, but how do you find out what the actual problem is?
>
> All the STOP messages I get are either UNHANDLED_KMODE_EXCEPTION or
> IRQL_NOT_LESS_OR_EQUAL. Could be anything.
Those are usually due to bad hardware drivers or bad hardware itself. Given
that you're seeing this on several machines, I'd place my bets on some bad
drivers that you've installed on all your machines. SOmetimes in the event
log you get a clue what the problem is, if not you can just use trial and
error, but it might take some time if you're only getting these errors once
a week.
> If there were anything I could do to find out, I'd do it. Unfortunately,
> there isn't.
Of course there is something you can do. How do you think other people sort
out problems like this? It is *not* normal behaviour, no matter what you
might think of MS.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
