 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Orchid XP v7 wrote:
>> Remote desktop. USB. Networking. 3D accelerated graphics. Memory
>> protection. Virtual memory. 64-bit OS support. Remote management.
>> Access control policies. Performance monitoring. Clustered drives.
>> RAID. Multi-OS boot. Support for about 63,417 different chunks of
>> hardware.
>
> OK then. Most of what you just listed is *hardware*, not *software*.
No it's not. For that which is hardware, add the phrase "support for" in
front, then.
> But, out of the box, there is no TCP/IP. So that's a
> useful thing that AmigaDOS doesn't have.
And there's a whole bunch more to "networking" than just "TCP/IP stack".
> Hardware support is... down to 3rd parties. If they write the drivers,
> you can use it. If they don't, you can't.
Not true. (1) look at how many of your device drivers on your Windows
box are authored by Microsoft. (2) The device drivers alone aren't
enough if there isn't a way of integrating them into the rest of the
system. (Granted, the Amiga had a fairly elegant way of integrating
things into the system.)
Consider, for example, the pain you went through on an Apple ][ to use
the disk drive card. And if Apple Basic didn't have a way to "print to
the card in slot N", even that wouldn't have worked. Nowadays, of
course, everyone does this.
Windows has to support all the infrastructure for loading and using and
unloading device drivers, in addition to the drivers. Yes, the Amiga had
that too, as do most modern OSes where you no longer recompile the
kernel to add devices. Just saying that the complexity there isn't 100%
in the drivers, and the more flexible it is for drivers, the more
compliated it is in the OS.
> Performance monitoring? I can do that.
You can? You can tell how many process switches per second are
happening, and whether you're slow because you're doing too many memory
allocations or too many disk reads or writes? You can map your graph of
SQL queries per seond on the same table as your graph of network
retransmissions per second to see if the SQL server is being slowed down
by that overhead? Cool.
> Oh, wait, aren't you the guy who
> claimed that AmigaDOS doesn't have premptive multitasking?
Never. You must have me confused with someone else.
> Multi-OS boot? Strange - I also run Debian Linux on my Amiga. :-P
Fair enough. I haven't really used an Amiga since the 1000/500 days. Who
is selling it nowadays?
> software should NOT crash just because
> you fed it bad data. Graceful failure, anyone?
Commercial software. Someone did a study
>>> Interesting. When I got my laptop, it crashed within 14 *seconds* of
>>> being turned on.
>>
>> One would think that would be either a misinstalled OS or a hardware
>> problem.
>
> If that were the case, visiting Windows Update a few times presumably
> wouldn't have fixed it...
If the file is corrupt on the disk, and a later update replaces that
file, I'd expect that would fix it, yah.
--
Darren New / San Diego, CA, USA (PST)
On what day did God create the body thetans?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Gilles Tran" <gil### [at] agroparistech fr> wrote in message
news:47b1bb80$1@news.povray.org...
>
> Microsoft is the *** 139th *** global corporation in terms of revenue,
> right after A.P. Moller-Maersk Group, which is a freight company, i.e.
> folks who use big boats to transport bananas around the globe.
> http://money.cnn.com/magazines/fortune/global500/2007/full_list/index.html
Is revenue the best measure? Or net-worth?
> That should keep Microsoft-themed hyperboles (and the relevance of the
> software industry compared to the banana shipping industry) in check I
> guess.
Fine, but we all know Microsoft is The Most Evil corporation ever to have
existed in recorded human history (aside from Disney). Then perhaps comes
Walmart, Exxon and Monsanto... And a Google search would probably back me
up. ;-)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Gail Shaw wrote:
> "There!" he says, pointing at the still executing query "See, it's done"
Heh. Reminds me of my first attempt to use Perl for CGI. I tried to run
it, and it didn't work. After a while I gave up waiting and closed IE. I
did some more poking and prodding of the code before the lead tech came
over and said "hey, you've using up 100% CPU on the webserver".
Oh. Oh yeah. Closing the window on your PC doesn't *actually* make the
server stop processing the request. Duh. :-S
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Orchid XP v7 wrote:
>
> Just being curios... What's a 'cursor', and why is that bad?
>
Cursors occupy resources on the server, and require the server to
maintain application state. The general understanding in n-tier systems
is the only thing that retains application state is the front-end
application, so resources on back-end servers can be released to other
clients.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
nemesis wrote:
> Darren New wrote:
>> Invisible wrote:
>>> Similarly, have you *ever* seen Linux crash?
>>
>> Regularly. 3 or 4 times last week alone. On a machine where Windows
>> never crashes. Sure, go ahead, blame the hardware.
>
> were you running as root?
Nope.
> desktop, server? task?
Desktop, editing code. Actually, I think firefox or thunderbird was
running. Altho it has also crashed while the screen saver was running.
The last couple times it was closing a page in firefox, IIRC.
And no, it's not just something doing a global grab or something. Cursor
locks up, and it doesn't answer pings.
--
Darren New / San Diego, CA, USA (PST)
On what day did God create the body thetans?
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Orchid XP v7" <voi### [at] devnull> wrote in message
news:47b1fefc$1@news.povray.org...
>
> Just being curios... What's a 'cursor', and why is that bad?
To quote a friend on SQL central. Cursors are RBAR (Row by agonising row)
processing
SQL's a set-based language. It's supposed to process datasets, not
individual rows. Here's a cursor syntax in T-SQL. My oracle's way too rusty
for me to attempt to do the same in pl-sql.
DECLARE curSomething CURSOR FOR
SELECT somecolumn FROM SomeTable
FETCH NEXT FROM curSomething INTO @AVariable
WHILE @@FETCH_STATUS = 0 -- there are rows in the cursor
BEGIN
-- Do something with the variable @AVariable
FETCH NEXT FROM curSomething INTO @AVariable
END
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Warp wrote:
>
> Much of IE's "performance" is deceiving.
Yes. But...
> For example, IE seems to take much less memory than other browsers
> because many of its components are system components already loaded
> into memory, and they are not listed under IE in the task manager.
> This means that IE is taking quite a lot of memory, but you won't see
> how much using task manager.
...plain Win+IE opened can still stay on pretty low memory usage. Try
plain Seamonkey (FF maybe too, haven't used for some time), reduce it
from Win+IE and notice that if IE did take same amount of memory, Win
(without IE) would get on with incredibly low memory.
--
Eero "Aero" Ahonen
http://www.zbxt.net
aer### [at] removethiszbxtnetinvalid
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Darren New wrote:
> Um, yes. So you're saying "Yes, MS made it possible to run the same
> executables on multiple operating systems over the course of decades.
> But that doesn't count, because it wasn't magic"?
no, I'm simply saying that those "multiple operating systems" are using
the same old code from the same single company to handle old apps, so
that doesn't count as multiple operating systems.
Multiple OSes in my book mean "Mac, Linux and Windows" rather than W95,
W98 or WNT.
I don't know how to make it any more clear so I'm stopping now.
>> What really made cross-platform software a little more closer to
>> reality were industry-strength standards,
>
> Like what? I haven't seen any standards that actually allow for programs
> to run cross-platform. Maybe I'm not following what you mean.
A pure Java/Python/whatever program using standard interprocess
communication channels to deal with the underlying OS and using a
cross-platform GUI such as Swing/GTK+ will run in pretty much any major
OS unaltered. Blender the graphical modeller has its interface
implemented in OpenGL, which means little problem porting it everywhere.
Coding a database using strict ISO SQL will make it pretty much
portable among many implementations.
> The internet doesn't make programs run cross-platform. It just gives you
> a way to access them when they're running on someone else's platform.
It doesn't really matter where a web app is running as long as it does
what I mean.
>> quite a few cross-platform development languages (Java, Python, Perl etc)
>
> And BASIC. Which used to run everywhere. And was promoted and
> implemented by MS (amongst others).
Visual Basic is not plain Basic. Heck, OpenOffice Basic is neither.
Yet, since the host app of OOBasic is cross-platform, OOBasic apps are
far more portable than VB.
> Note that Java, Python, etc are all libraries controlled by one company
> or individual that give you cross-platform APIs for your code. Just like
> Win32.
bzzt. wrong! They have Foundations behind them and many, many
individuals from different places moving them forward by meritocracy.
Everyone can apply to contribute to their development, including
individuals from Microsoft.
> Except they don't provide executables.
They sure do. And they are truly cross-platform.
> I agree that open source software is probably a better way to do stuff,
> as long as you aren't worried about making a living at it.
the folks at RedHat or Novell seem to be doing fine.
> Uh, so, how much Linux software runs without recompiling under Solaris?
> Does that actually work?
no, different executable format. One of the areas that are not
standardized. But the source is there and autoconf makes sure the right
compilation flags for your system is used. Or just grab it from the
binary repositories of modern Linux distros...
> I think the only thing that's actually going to satisfy you is virtual
> machine stuff, like VMWare.
no, because that wold require Windows to be running underneath as well.
If you want truly cross-platform apps without recompiling, target a
programming language with a cross-platform runtime. You won't need to
recompile your apps, but should install the proper runtime for your
system, which will be provided from the people who compiled them to your
system.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Darren New wrote:
>> But, out of the box, there is no TCP/IP. So that's a useful thing that
>> AmigaDOS doesn't have.
>
> And there's a whole bunch more to "networking" than just "TCP/IP stack".
OK. I'll grant you that one.
>> Hardware support is... down to 3rd parties. If they write the drivers,
>> you can use it. If they don't, you can't.
>
> Not true. (1) look at how many of your device drivers on your Windows
> box are authored by Microsoft.
I didn't realise there were *any*? [I mean, beyond the generic VGA
driver, serial port driver, etc.]
> Just saying that the complexity there isn't 100%
> in the drivers, and the more flexible it is for drivers, the more
> compliated it is in the OS.
OK. Fair enough.
>> Performance monitoring? I can do that.
>
> You can? You can tell how many process switches per second are
> happening, and whether you're slow because you're doing too many memory
> allocations or too many disk reads or writes? You can map your graph of
> SQL queries per seond on the same table as your graph of network
> retransmissions per second to see if the SQL server is being slowed down
> by that overhead? Cool.
Actually no.
What I can do is get a graph of various things like number of processes
running, CPU usage per process, memory per process, memory free, etc.
Integrating that with performance data from other apps? Well, that would
require the other apps to supply performance data. And some ARexx
scripting to link them together. And by the time you get there, you
might as well write the monitoring app yourself, at which point it
ceases to be about what the "OS" can do.
AmigaDOS gives you the system calls to get performance stats. There is
software that can graph it for you. That I can say.
>> Oh, wait, aren't you the guy who claimed that AmigaDOS doesn't have
>> premptive multitasking?
>
> Never. You must have me confused with someone else.
Oh good, that's a start... :-}
>> Multi-OS boot? Strange - I also run Debian Linux on my Amiga. :-P
>
> Fair enough. I haven't really used an Amiga since the 1000/500 days. Who
> is selling it nowadays?
Selling it? Heh. Last I heard [long time ago], "Amiga" is a
software-only company now. No, I have no idea how the hell that makes
any sense... IIRC, AmigaOS was ported to PPC or something. Whatever.
The hardware hasn't been on sale for a while now, AFAIK. But then, would
*you* buy a "personal computer" powered by a 20 MHz 68030 with 2 MB RAM?
>>> One would think that would be either a misinstalled OS or a hardware
>>> problem.
>>
>> If that were the case, visiting Windows Update a few times presumably
>> wouldn't have fixed it...
>
> If the file is corrupt on the disk, and a later update replaces that
> file, I'd expect that would fix it, yah.
Mmm, I guess. However, given how much everybody complained about XP
crashing when it first came out, and it's not *as bad* today, I think
it's a case of M$ actually fixing some of the more serious bugs...
--
http://blog.orphi.me.uk/
http://www.zazzle.com/MathematicalOrchid*
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Darren New wrote:
>
> "It is not so much user friendly as it is expert tolerant."
>
Unfortunately it's becoming TOO "user-friendly" and popular. I think
I'll start looking for *BSD's next, to stay in front of the
hey-you-know-so-you-can-tell-me-why-i'm-stupid -people. Most of them
realize I don't want (or "know" or really know :p) to help them with
some system I don't personally like.
:)
--
Eero "Aero" Ahonen
http://www.zbxt.net
aer### [at] removethiszbxtnetinvalid
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
