On 16/04/2012 13:33, Robin wrote:
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.


Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff. But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

It's not so much compilers that care, it's people. Use of goto can make
your code a nightmare to read.

Robin :
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.

Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff.

It's not the compiler that's the problem, it's the maintenance
programmer who has to pick up the program and understand what it does,
and how.

But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

I think in recent decades programmers have changed, such that they're
now much less likely to reach for the GOTO statement even if it is
available.

--
Mike Barnes

On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.

It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc

Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?

Exists in C too, but in both languages you don't need to actually use it.

If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.

And "C" would also be a very poor language for teaching too.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

On 16/04/12 13:33, Robin wrote:

Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff. But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

The structure in program code (ie "GOTO considered harmful") is there
for the benefit of the human reader. At the very lowest level it is
little more than compare and branch all te way down. But these days
nobody goes the processors and microcode are designed to work with
compilers not hand written machine codes,



--
djc

On 16/04/2012 13:58, Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.

It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc

Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?

Exists in C too, but in both languages you don't need to actually use it.

If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.

Part of learning is that sometimes things are provided which you
shouldn't use.

CF DIY and angle grinders...

On Mon, 16 Apr 2012 14:18:47 +0100, Clive George
wrote:

On 16/04/2012 13:58, Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.

It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc

Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?

Exists in C too, but in both languages you don't need to actually use it.

If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.

Part of learning is that sometimes things are provided which you
shouldn't use.

Yes. However I wouldn't first recommend making good practice a habit
by restricting the availability of "things you shouldn't use"

CF DIY and angle grinders...

I also would not recomend first starting DIY with an angle grinder.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

"Adrian C" wrote in message
...
On 16/04/2012 12:27, dennis@home wrote:

Everyone except you seems to understand...

I understand.
A POP client downloads stuff to a local mail file.
Typically the server is somewhere out there on the net.
The local files are on the machine the client is on.
The servers mail database is not a local file to the POP client at all.
You can argue what you like but it doesn't change the facts.

You know little about _servers_ and the connections they make.

I know that the whole thing is about TNP saying "no POP client" and not "no
POP server".
however you can continue to wriggle about servers for as long as you like.

"Tim Streater" wrote in message
...
In article ,
"dennis@home" wrote:

I understand.
A POP client downloads stuff to a local mail file.
Typically the server is somewhere out there on the net.
The local files are on the machine the client is on.
The servers mail database is not a local file to the POP client at all.
You can argue what you like but it doesn't change the facts.

I think everyone knew this but you appeared to be arguing the opposite.

No i am still arguing what started this argument while you are arguing about
something else.

Robin wrote

VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.

Does the use of GOTO matter much - or at all - with modern compilers?

The main problem with GOTOs is that the make the code very hard to read
when used excessively. The problem isnt with how efficiently the code runs.

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff.

That isnt really feasible with the worst of the mess
that the extreme use of GOTOs can produce.

But pl bear in mind that for me "modern" is anything after the mid-70s ;(

"John Rumm" wrote in message
o.uk...
On 16/04/2012 12:27, dennis@home wrote:


"John Rumm" wrote in message
o.uk...
On 15/04/2012 22:19, dennis@home wrote:


"John Rumm" wrote in message
o.uk...
On 15/04/2012 20:48, dennis@home wrote:


"Adrian C" wrote in message
...
On 15/04/2012 20:34, Adrian C wrote:
On 15/04/2012 19:15, dennis@home prattled:

"The Natural Philosopher" wrote in message


I had to evaluate it for IP use when it first appeared. As it was
essentially X-400 with a bit of TCP/IP and SMTP gatewaying
bolted on
the front,.

I was actually quite impressed.

OK it ran on NT and it wasn't a patch on sendmail. But compared
with
microsoft mail it was a huge step forward. Very corporate
though. No
pop clients to a local mail folder. Oh no. Cant have that!

Anyway read what you have quoted "very corporate" ,"no pop clients",
"can't have that".
Well Outlook is a POP client and guess what its a M$ corporate product.

Get a grip you dozey twonk, he was talking about exchange when it
first came out.

What I said was true and relevant even if you didn't get it.

True, yes: outlook does pop. Relevant, not at all, since we were
talking about the release version of exchange and not about outlook.

Most of them will IME. Hence they can gateway from external mail
stores. (files local to the server that is)

So not local files then.

Yes, think about it. If the files are stored on the same machine that
is retrieving the messages from a remote sever, then they are
definitely local.

It isn't even likely to have access to local files.
Local as living on a server doesn't make much sense.

Of course it does. Exchange server sucks email from other servers,
stores it locally in its DB, and then hands it out to its clients on
demand. Not rocket science is it?

So now you want to believe local files and databases are actually the

Well given that true random access and hashed FB systems that access a
disk without also allocating the used blocks into a file are these
days very rare, I think its safe to say yes.

same thing and have the meaning TNP implied..

Everyone except you seems to understand...

I understand.

And yet everything you say suggests otherwise...

A POP client downloads stuff to a local mail file.

Agreed - often it will (although often the "local" mail file will be
mapped to a network drive, but lets ignore that to keep it simple for
you).

Typically the server is somewhere out there on the net.

You are not thinking corporate - i.e. the typical market for products like
Exchange sever. That often resides on your corporate intranet.

The local files are on the machine the client is on.

In the case of a mail client yes. So what about when the Exchange server
is also the client? The same is true in many cases. Exchange is often
configured as an email proxy. It reaches out to other internet based
servers as a client, to retrieve mail and store it locally. Later client
machines on the server's intranet use it to access email.

The servers mail database is not a local file to the POP client at all.

No one was claiming it was. Its local to the *corporate* exchange server
in many cases.

Then why are you arguing?


You can argue what you like but it doesn't change the facts.

I have no need to change the facts, I was just attempting to improve your
understanding of them.

I understand fully, you are arguing for the sake of it.


Have you worked out the difference between a LED source and a filament
source yet?

I can buy them both from the same source...? what are you prattling about
now?

Still wriggling I see.

Mike Barnes wrote:
The Natural Philosopher :
Mike Barnes wrote:
Robin :
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.
Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff.
It's not the compiler that's the problem, it's the maintenance
programmer who has to pick up the program and understand what it does,
and how.

But
pl bear in mind that for me "modern" is anything after the mid-70s ;(
I think in recent decades programmers have changed, such that
they're
now much less likely to reach for the GOTO statement even if it is
available.

But you cant write a switch() statement without a 'break' really and
that is the same thing:

Well if you choose to view it as "the same thing" that's entirely up to
you. But I don't think of it that way, any more than I think of an "if"
statement as "the same thing" as a "goto".



Both include unconditional jumps to target locations



I.e.

IF NOT THIS GOTO ELSE:
DO THIS
GO TO ENDIF
ELSE :
DO THE OTHER
ENDIF:

is exactly the same as
IF THIS
THEN THAT
ELSE
DO THE OTHER
ENDIF.

And in fact its how the processor machine code generally comes out.

likewise a

SWITCH(THIS)
case: A
DO A
break

ENDSWITCH


is essentially
IF (THIS==A)
DO A
GOTO ENDSWITCH.


in C terms the closing curly brace is identified as the unconditional
goto target, that's all.


Or in the case of a CONTINUE statement in a loop, the BEGINNING curly brace.

As the END curly brake in a FOR loop or a WHILE loop is in itself a GOTO
to the beginning of the loop.


All these language constructions are simply ways of having gotos without
actually writing a goto.

So there cannot be anything WRONG with a GOTO in PRINCIPLE because all
languages and all processors have them and need them.

So it becomes merely an issue of readability and style.

And my simple point is that someimes NOT using one create turgid
unreadable code.


If you have multiple nests of if then else constructs and somewhere
inside the 13th 'russian doll' set of braces you need to get out to
somewhere else entirely a Goto is a damn sight easier to follow than 12
braces and a further test.







--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

John Rumm wrote:
On 16/04/2012 16:34, The Natural Philosopher wrote:
Robin wrote:
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.


Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff. But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

GOTO is a valid construct. The problem is when you uses it top produce
spaghetti code.


Any IF...THEN...ELSE... will compile dowmn to at least one unconditional
jump.

C supports not only 'goto' but 'break' and 'continue' which are
specialised versions of the same thing.


As well as setjmp() and longjmp() which are even more flagrant abusers
of structure.

But thunderingly useful when you are 40 subroutines deep and the
hardware says 'er...I just lost carrier' and you want to simply abort
everything and start over ...without returning the same error code up
thorough the 40 subroutines.

Indeed, although these days one would argue that's what proper exception
handling is for...


and how else is that to be arranged?





--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

John Rumm wrote:
On 16/04/2012 16:39, The Natural Philosopher wrote:
Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.
It's OT for this thread, but for your information BBC BASIC is
currently
one of the most popular languages for teaching programming in UK
schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc


Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?
Exists in C too, but in both languages you don't need to actually use
it.

If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.


The point is that sometimes goto IS good practice.

Its a fine way to make a simple clear direct statement for error
handling and stuff that cuts across - and has to cut across - the normal
flow of programming.

GOTO jail. Go directly to jail, Do NOT pass go. Do NOT collect £200.


And "C" would also be a very poor language for teaching too.

Obviously anything that is pragmatically useful and written by engineers
to actually write real life code on is useless for academics who don't
want to sully their hands with the real world,

Although you have to admit, that if said engineers had had decent
interactive terminals with screen editors etc, they may not have
insisted on something quit as concise and dependent on using every
printable character in the ASCII set for something! ;-)

I am not sure they used the ¬ symbol anywhere. :-) or the `

I do remember a friend gave up on C because his AppleII didnt have {} on
the keyboard.

In the end he had a #define BEGIN { and #define END } or something like
that set up..



--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

"The Natural Philosopher" wrote in message
...

For everything else there is Linux. Massively robust and functional.
Thouugh it did go doolally trying to transer a 15 Mbyte file via NFS to a
remote machines on the net...NFS is a fine protocol, but not over links
that occasioally do down..:-)


NFS is an awful protocol.
It doesn't keep state.
Other processes can open the same file and write bits to it without the
other processes being aware.
That is why so many programs use stupid lockfiles to give some attempt and
stopping files with part data from other processes being created.
It works as long as all the processes are written to look for the lockfile
but it sure ain't a good protocol.

george [dicegeorge] wrote:
On 16/04/12 16:48, The Natural Philosopher wrote:
Mark wrote:
On Thu, 12 Apr 2012 12:38:29 +0100, John Rumm
wrote:

[-snip-]

However, XP mode is in reality a complete virtual machine running a
real copy of WinXP. There is nothing to stop you using any other
virtual PC hypervisor (including Microsoft's own Virtual PC) and
installing your own real copy of XP on that.

Running virtual machines can be very useful. However I would avoid
Microsoft Virtual PC. It's v-e-r-y slow. Oracle (Sun) VirtualBox is
much better IMHO.


Virtualbox is - adequate.

In order to get stable operation I have had to turn off MOST of the
'hardware acceleration' and that makes cursor response a but crappy, and
also flash videos don't really run smoothly under my XP machine.

But that's OK because of what I want my windows FOR. which is graphics
work.

Its useless for all but the simplest games, but its fine for RhinoCAD
and CorelDraw. And by doing what amounts to a suspend rather than a
shutdown, I can 'reboot' it in a few seconds.

For everything else there is Linux. Massively robust and functional.
Thouugh it did go doolally trying to transer a 15 Mbyte file via NFS to
a remote machines on the net...NFS is a fine protocol, but not over
links that occasioally do down..:-)




drat - virtualbox only 'adequate'?
I'm buying a new laptop,
my plan was to have Windows7 then virtual box then Xubuntu,
and I'd thought it would run almost as fast as if dualbooted.
Where is your slowdown coming from?


Video layer

simulating a video card + driver that maps to a windowed X windows is
not massively fast..

everything else is fine: in program computation gets pretty much as much
CPU as the basic linux timeslice allows, and its not a huge overhead in
context switching, so CPU intensive stuff is fine.

networking takes a BIT of a hit - again there is a 'virtual ethernet
card' presented.


But its for sure fast enough to access the internet full speed, even if
the 100Mbps LAN is only about 20...

So whatever you play games on/watch HD TV over the internet on /watch
flash on, make that your primary OS.


I am not sure why one would want a virtual Linux machine tho. Except to
play with.


I had playing with computers in about 1984 when I started to program
them for a living..

What I want is a stable platform for the things I need to do, and thats
linux. I tolerate windows because I cat get two apps I need on any other OS.




[george]


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

Tim Streater wrote:
In article ,
Huge wrote:

On 2012-04-16, Tim Streater wrote:
In article ,
John Rumm wrote:


Obviously anything that is pragmatically useful and written by
engineers
to actually write real life code on is useless for academics who
don't
want to sully their hands with the real world,
Although you have to admit, that if said engineers had had
decent interactive terminals with screen editors etc, they may not
have insisted on something quit as concise and dependent on using
every printable character in the ASCII set for something! ;-)

Not sure about that. Remember, these are the ****ers that gave us
TECO. And two-character unix commands.

I think describing them as "****ers" is a little strong. Such
language should
be reserved for those, er, ****ers in Redmond.

From time to time I get dopey enough to think that I ought to learn the
ins and outs of unix - environment variables, shell scripting, edlin,
emacs, etc etc.


I don't use any of those.

script is for monkeys who cant write C :-)

Every time I try a regexp it blows up on me and I write some C code that
does EXACTLY what I want, not what some **** decided would be useful for
me to do.

I program with a PROPER editor (Geany) in C and PHP.
Or write docs in Open Office.

After five minutes (or twenty [1] if I had been fool enough to start
emacs), I feel like throwing the computer out of the window. Then I'm
sane again until the next time. Luckily these episodes are becoming
rarer as I get older.

[1] actually it's five minutes in all cases, it's just that it takes an
extra fifteen minutes to figure out how to quit emacs.


well exactly.

The beauty of - say - a modern Gnome GUI is it behaves pretty much like
windows XXP except its faster and never crashes. Geekls may sniff at
that, bit I find it makes things easier.

Click on an editor icon, there's an editor. Nuff said.

I admit to occasionally using vi to edit root level files, but thats cos
I HAD to learn it as the ONLY editor that was ALWAYS present on ANY *nix
system.

And I avoid editing root level files except as a last resort.

The fact is you can install linux and get networking browsing email
office (word/calc/powerpoint) plus a media player that will play DVDS
all as a standard issue distro without ever having to edit anything at all.

And if that is your basic toolkit, there's no need top step outside the GUI.


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

dennis@home wrote:


"Adrian C" wrote in message
...
On 16/04/2012 12:27, dennis@home wrote:

Everyone except you seems to understand...

I understand.
A POP client downloads stuff to a local mail file.
Typically the server is somewhere out there on the net.
The local files are on the machine the client is on.
The servers mail database is not a local file to the POP client at all.
You can argue what you like but it doesn't change the facts.

You know little about _servers_ and the connections they make.

I know that the whole thing is about TNP saying "no POP client" and not
"no POP server".

which you failed to understand in practice amounts to the same thing: a
pop client wont work without a pop server will it?


So you could not e.g. run a Eudora client against an exchange server and
download your mail from it to a local file system on the client machine.


Now I could have spelt that out in words of one syllable especially for
thick ****s like you, but frankly everyone else got it and to be honest,
where you are concerned, I am not sure I really give a ****.



however you can continue to wriggle about servers for as long as you like.


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

dennis@home wrote:


"The Natural Philosopher" wrote in message
...

For everything else there is Linux. Massively robust and functional.
Thouugh it did go doolally trying to transer a 15 Mbyte file via NFS
to a remote machines on the net...NFS is a fine protocol, but not over
links that occasioally do down..:-)


NFS is an awful protocol.
It doesn't keep state.
Other processes can open the same file and write bits to it without the
other processes being aware.
That is why so many programs use stupid lockfiles to give some attempt
and stopping files with part data from other processes being created.
It works as long as all the processes are written to look for the
lockfile but it sure ain't a good protocol.


Oh dear.
No. I am not going to say how stupid teh above is.

I simply cant be arsed.


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

"The Natural Philosopher" wrote in message
...


I do remember a friend gave up on C because his AppleII didnt have {} on
the keyboard.

In the end he had a #define BEGIN { and #define END } or something like
that set up..

You just did a global find and replace.

Tim Streater wrote:
In article ,
The Natural Philosopher wrote:

Tim Streater wrote:

From time to time I get dopey enough to think that I ought to learn
the ins and outs of unix - environment variables, shell scripting,
edlin, emacs, etc etc.

I don't use any of those.

Well neither do I.

script is for monkeys who cant write C :-)

I can't be arsed to write C any longer, so I do it all in PHP, including
CLI scripts.

Every time I try a regexp it blows up on me and I write some C code
that does EXACTLY what I want, not what some **** decided would be
useful for me to do.

regexps are line noise, by and large. As are TECO commands.

I program with a PROPER editor (Geany) in C and PHP.

TextWrangler under OS X.

Or write docs in Open Office.

After five minutes (or twenty [1] if I had been fool enough to start
emacs), I feel like throwing the computer out of the window. Then
I'm sane again until the next time. Luckily these episodes are
becoming rarer as I get older.
[1] actually it's five minutes in all cases, it's just that it
takes an extra fifteen minutes to figure out how to quit emacs.

well exactly.

The beauty of - say - a modern Gnome GUI is it behaves pretty much
like windows XXP except its faster and never crashes. Geekls may sniff
at that, bit I find it makes things easier.

Click on an editor icon, there's an editor. Nuff said.

I admit to occasionally using vi to edit root level files, but thats
cos I HAD to learn it as the ONLY editor that was ALWAYS present on
ANY *nix system.

Yep yep, I still have (actually two, for some reason) an Ultrix cheat
card for vi. But even vi has a 127-page manual that you can pay dosh
for. I've never understood why an editor should need a manual.

And I avoid editing root level files except as a last resort.

The fact is you can install linux and get networking browsing email
office (word/calc/powerpoint) plus a media player that will play DVDS
all as a standard issue distro without ever having to edit anything at
all.

And if that is your basic toolkit, there's no need top step outside
the GUI.

OS X in my case. You can also install Xcode, which I found to be a handy
way of just clicking to get the occasional C program I want to alter, to
compile/build. Such as the sqlite shell, or vasm and vlink (for
cross-assembling 68000 code). That way, I can completely ignore the
existence of 'make'.

yeah. I frigged with OSX but frankly I didn't like it.
Cant afford a decent mac..and its enough keeping my wifes G5 going
through the inevitable crashes

I thin the final nail in that coffin was when a friend brought his late
sons I phone over 'to get some pictures off'

The mac didn't recognise it. G5 wont run latest code... Linux did tho.

And yet the hardware is of comparable age.



--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

"The Natural Philosopher" wrote in message
...
dennis@home wrote:


"Adrian C" wrote in message
...
On 16/04/2012 12:27, dennis@home wrote:

Everyone except you seems to understand...

I understand.
A POP client downloads stuff to a local mail file.
Typically the server is somewhere out there on the net.
The local files are on the machine the client is on.
The servers mail database is not a local file to the POP client at all.
You can argue what you like but it doesn't change the facts.

You know little about _servers_ and the connections they make.

I know that the whole thing is about TNP saying "no POP client" and not
"no POP server".

which you failed to understand in practice amounts to the same thing: a
pop client wont work without a pop server will it?

Rubbish, if you wanted to say no POP server then you should have said no POP
server and not argued about what you said.
It was easy enough to correct what you said and not argue.



So you could not e.g. run a Eudora client against an exchange server and
download your mail from it to a local file system on the client machine.

But that isn't what you said.



Now I could have spelt that out in words of one syllable especially for
thick ****s like you, but frankly everyone else got it and to be honest,
where you are concerned, I am not sure I really give a ****.

I am not sure that you could use the correct one syllable words.
You can't manage it with two or three and there is no evidence you can
manage one.

"The Natural Philosopher" wrote in message
...
dennis@home wrote:


"The Natural Philosopher" wrote in message
...

For everything else there is Linux. Massively robust and functional.
Thouugh it did go doolally trying to transer a 15 Mbyte file via NFS to
a remote machines on the net...NFS is a fine protocol, but not over
links that occasioally do down..:-)


NFS is an awful protocol.
It doesn't keep state.
Other processes can open the same file and write bits to it without the
other processes being aware.
That is why so many programs use stupid lockfiles to give some attempt
and stopping files with part data from other processes being created.
It works as long as all the processes are written to look for the
lockfile but it sure ain't a good protocol.


Oh dear.
No. I am not going to say how stupid teh above is.

Well yes I have said how stupid it is.
Why don't you say why you think its wrong so we can have another worthless
argument where you fail to be correct.


I simply cant be arsed.

Huge wrote:

On 2012-04-16, Tim Streater wrote:
In article ,
John Rumm wrote:

That was said to be the mark of a guru - he could accurately predict
what actions would happen if he typed his name into TECO in command mode.

Mmm. IMO, gurus need to be taken out and shot, for the most part.

Oi! Some of us make a comfy living out of guru-hood.

Ommmmmmmmmm.....

On 16/04/2012 21:48, dennis@home wrote:


"John Rumm" wrote in message
o.uk...
On 16/04/2012 12:27, dennis@home wrote:


"John Rumm" wrote in message
o.uk...
On 15/04/2012 22:19, dennis@home wrote:


"John Rumm" wrote in message
o.uk...
On 15/04/2012 20:48, dennis@home wrote:


"Adrian C" wrote in message
...
On 15/04/2012 20:34, Adrian C wrote:
On 15/04/2012 19:15, dennis@home prattled:

"The Natural Philosopher" wrote in message


I had to evaluate it for IP use when it first appeared. As it
was
essentially X-400 with a bit of TCP/IP and SMTP gatewaying
bolted on
the front,.

I was actually quite impressed.

OK it ran on NT and it wasn't a patch on sendmail. But compared
with
microsoft mail it was a huge step forward. Very corporate
though. No
pop clients to a local mail folder. Oh no. Cant have that!

Anyway read what you have quoted "very corporate" ,"no pop clients",
"can't have that".
Well Outlook is a POP client and guess what its a M$ corporate
product.

Get a grip you dozey twonk, he was talking about exchange when it
first came out.

What I said was true and relevant even if you didn't get it.

True, yes: outlook does pop. Relevant, not at all, since we were
talking about the release version of exchange and not about outlook.

Most of them will IME. Hence they can gateway from external mail
stores. (files local to the server that is)

So not local files then.

Yes, think about it. If the files are stored on the same machine that
is retrieving the messages from a remote sever, then they are
definitely local.

It isn't even likely to have access to local files.
Local as living on a server doesn't make much sense.

Of course it does. Exchange server sucks email from other servers,
stores it locally in its DB, and then hands it out to its clients on
demand. Not rocket science is it?

So now you want to believe local files and databases are actually the

Well given that true random access and hashed FB systems that access a
disk without also allocating the used blocks into a file are these
days very rare, I think its safe to say yes.

same thing and have the meaning TNP implied..

Everyone except you seems to understand...

I understand.

And yet everything you say suggests otherwise...

A POP client downloads stuff to a local mail file.

Agreed - often it will (although often the "local" mail file will be
mapped to a network drive, but lets ignore that to keep it simple for
you).

Typically the server is somewhere out there on the net.

You are not thinking corporate - i.e. the typical market for products
like Exchange sever. That often resides on your corporate intranet.

The local files are on the machine the client is on.

In the case of a mail client yes. So what about when the Exchange
server is also the client? The same is true in many cases. Exchange is
often configured as an email proxy. It reaches out to other internet
based servers as a client, to retrieve mail and store it locally.
Later client machines on the server's intranet use it to access email.

The servers mail database is not a local file to the POP client at all.

No one was claiming it was. Its local to the *corporate* exchange
server in many cases.

Then why are you arguing?

Because you were going on about Outlook which was not particularly
helpful in a discussion about Exchange.

Remember you asked "Since when did mail servers do POP to local files?",
so I told you when Exchange will do POP to local files.

You can argue what you like but it doesn't change the facts.

I have no need to change the facts, I was just attempting to improve
your understanding of them.

I understand fully,

If that were the case, you would not have asked the question above...

you are arguing for the sake of it.

The irony...

Have you worked out the difference between a LED source and a filament
source yet?

I can buy them both from the same source...? what are you prattling
about now?

Still wriggling I see.

I really have no idea what you are talking about.

I presume your senility is getting to you and you are talking about some
other thread I have not even contributed to...


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On Mon, 16 Apr 2012 22:20:53 +0100, Tim Streater wrote:

In article ,
Bob Eager wrote:

On Mon, 16 Apr 2012 18:45:34 +0100, Tim Streater wrote:

In article ,
Bob Eager wrote:

On Mon, 16 Apr 2012 17:47:14 +0100, Tim Streater wrote:

Not sure about that. Remember, these are the ****ers that gave us
TECO. And two-character unix commands.

Nothing wrong with TECO! I met someone who had it solving
differential equations, too!

So which key did you press for 'undo', then, Our Bob?

I was so good, I never needed to.

*That* good, eh? Have another gin.

I decided to learn TECO rather than the locally written editor because I
thought it would be useful in the long run.

And it was. Multiple DEC systems (RT-11, RSX, VMS, etc.) plus - strangely
- the assembler ROM on the BBC B.



--
Use the BIG mirror service in the UK:
http://www.mirrorservice.org

*lightning protection* - a w_tom conductor

On 16/04/2012 21:50, The Natural Philosopher wrote:
John Rumm wrote:
On 16/04/2012 16:34, The Natural Philosopher wrote:
Robin wrote:
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.


Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff. But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

GOTO is a valid construct. The problem is when you uses it top produce
spaghetti code.


Any IF...THEN...ELSE... will compile dowmn to at least one unconditional
jump.

C supports not only 'goto' but 'break' and 'continue' which are
specialised versions of the same thing.


As well as setjmp() and longjmp() which are even more flagrant abusers
of structure.

But thunderingly useful when you are 40 subroutines deep and the
hardware says 'er...I just lost carrier' and you want to simply abort
everything and start over ...without returning the same error code up
thorough the 40 subroutines.

Indeed, although these days one would argue that's what proper
exception handling is for...


and how else is that to be arranged?

I mean in the specific sense of languages that actually have exception
handling constructs built in, to support throwing and catching
exceptions and all that entails[1], rather than just the generic sense
of how you handle things programatically when the world does not go the
way one would like.

[1] Like automatically invoking destructors on the hierarchy of objects
up the call chain (even when some of those methods are late bound
virtual methods that may have not even existed when the exception
handling was originally written), making sure "finally" blocks are
executed, that the stack is tidied up, exception objects are caught in
places able to handle them, and their payload can be used etc...

Some of these requirements are specific in nature to OO designs, and
hence benefit greatly from specific capabilities added the languages
designed to handle the complexities introduced.


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

John Rumm wrote:
On 16/04/2012 21:50, The Natural Philosopher wrote:
John Rumm wrote:
On 16/04/2012 16:34, The Natural Philosopher wrote:
Robin wrote:
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.


Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into control
structures so they didn't care if you used goto or cleverer stuff. But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

GOTO is a valid construct. The problem is when you uses it top produce
spaghetti code.


Any IF...THEN...ELSE... will compile dowmn to at least one
unconditional
jump.

C supports not only 'goto' but 'break' and 'continue' which are
specialised versions of the same thing.


As well as setjmp() and longjmp() which are even more flagrant abusers
of structure.

But thunderingly useful when you are 40 subroutines deep and the
hardware says 'er...I just lost carrier' and you want to simply abort
everything and start over ...without returning the same error code up
thorough the 40 subroutines.

Indeed, although these days one would argue that's what proper
exception handling is for...


and how else is that to be arranged?

I mean in the specific sense of languages that actually have exception
handling constructs built in, to support throwing and catching
exceptions and all that entails[1], rather than just the generic sense
of how you handle things programatically when the world does not go the
way one would like.

[1] Like automatically invoking destructors on the hierarchy of objects
up the call chain (even when some of those methods are late bound
virtual methods that may have not even existed when the exception
handling was originally written), making sure "finally" blocks are
executed, that the stack is tidied up, exception objects are caught in
places able to handle them, and their payload can be used etc...

Some of these requirements are specific in nature to OO designs, and
hence benefit greatly from specific capabilities added the languages
designed to handle the complexities introduced.


well I bet whoever wrote the compiler used some pretty arcane components
then :-)


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

On 17/04/2012 00:13, Bob Eager wrote:
On Mon, 16 Apr 2012 22:20:53 +0100, Tim Streater wrote:

In ,
Bob wrote:

On Mon, 16 Apr 2012 18:45:34 +0100, Tim Streater wrote:

In ,
Bob wrote:

On Mon, 16 Apr 2012 17:47:14 +0100, Tim Streater wrote:

Not sure about that. Remember, these are the ****ers that gave us
TECO. And two-character unix commands.

Nothing wrong with TECO! I met someone who had it solving
differential equations, too!

So which key did you press for 'undo', then, Our Bob?

I was so good, I never needed to.

*That* good, eh? Have another gin.

I decided to learn TECO rather than the locally written editor because I
thought it would be useful in the long run.

And it was. Multiple DEC systems (RT-11, RSX, VMS, etc.) plus - strangely
- the assembler ROM on the BBC B.

More of an edit/tpu man myself ;-)

--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

"The Natural Philosopher" wrote in message
...
John Rumm wrote:
On 16/04/2012 21:50, The Natural Philosopher wrote:
John Rumm wrote:
On 16/04/2012 16:34, The Natural Philosopher wrote:
Robin wrote:
VB (and BBC Basic) have procedual and loop statements, that if used
properly negate the need to scatter GOTO statements around the
source. I can write a program in VB that would be structured
identically in C (or whatever) just replacing little bits. No
references to line numbers required.


Does the use of GOTO matter much - or at all - with modern compilers?

I ask as I thought that modern compilers broke stuff down into
control
structures so they didn't care if you used goto or cleverer stuff.
But
pl bear in mind that for me "modern" is anything after the mid-70s ;(

GOTO is a valid construct. The problem is when you uses it top produce
spaghetti code.


Any IF...THEN...ELSE... will compile dowmn to at least one
unconditional
jump.

C supports not only 'goto' but 'break' and 'continue' which are
specialised versions of the same thing.


As well as setjmp() and longjmp() which are even more flagrant abusers
of structure.

But thunderingly useful when you are 40 subroutines deep and the
hardware says 'er...I just lost carrier' and you want to simply abort
everything and start over ...without returning the same error code up
thorough the 40 subroutines.

Indeed, although these days one would argue that's what proper
exception handling is for...


and how else is that to be arranged?

I mean in the specific sense of languages that actually have exception
handling constructs built in, to support throwing and catching exceptions
and all that entails[1], rather than just the generic sense of how you
handle things programatically when the world does not go the way one
would like.

[1] Like automatically invoking destructors on the hierarchy of objects
up the call chain (even when some of those methods are late bound virtual
methods that may have not even existed when the exception handling was
originally written), making sure "finally" blocks are executed, that the
stack is tidied up, exception objects are caught in places able to handle
them, and their payload can be used etc...

Some of these requirements are specific in nature to OO designs, and
hence benefit greatly from specific capabilities added the languages
designed to handle the complexities introduced.


well I bet whoever wrote the compiler used some pretty arcane components
then :-)

I'll bet our roddy has a few things to say about this.
LMFAO.
uk.d-i-y awaits (with baited breath) our bull****ter of the year.
The stage is yours, rodders.

On Mon, 16 Apr 2012 22:00:48 +0100, The Natural Philosopher
wrote:

george [dicegeorge] wrote:
On 16/04/12 16:48, The Natural Philosopher wrote:
Mark wrote:
On Thu, 12 Apr 2012 12:38:29 +0100, John Rumm
wrote:

[-snip-]

However, XP mode is in reality a complete virtual machine running a
real copy of WinXP. There is nothing to stop you using any other
virtual PC hypervisor (including Microsoft's own Virtual PC) and
installing your own real copy of XP on that.

Running virtual machines can be very useful. However I would avoid
Microsoft Virtual PC. It's v-e-r-y slow. Oracle (Sun) VirtualBox is
much better IMHO.


Virtualbox is - adequate.

In order to get stable operation I have had to turn off MOST of the
'hardware acceleration' and that makes cursor response a but crappy, and
also flash videos don't really run smoothly under my XP machine.

But that's OK because of what I want my windows FOR. which is graphics
work.

Its useless for all but the simplest games, but its fine for RhinoCAD
and CorelDraw. And by doing what amounts to a suspend rather than a
shutdown, I can 'reboot' it in a few seconds.

For everything else there is Linux. Massively robust and functional.
Thouugh it did go doolally trying to transer a 15 Mbyte file via NFS to
a remote machines on the net...NFS is a fine protocol, but not over
links that occasioally do down..:-)




drat - virtualbox only 'adequate'?
I'm buying a new laptop,
my plan was to have Windows7 then virtual box then Xubuntu,
and I'd thought it would run almost as fast as if dualbooted.
Where is your slowdown coming from?


Video layer

simulating a video card + driver that maps to a windowed X windows is
not massively fast..

everything else is fine: in program computation gets pretty much as much
CPU as the basic linux timeslice allows, and its not a huge overhead in
context switching, so CPU intensive stuff is fine.

networking takes a BIT of a hit - again there is a 'virtual ethernet
card' presented.


But its for sure fast enough to access the internet full speed, even if
the 100Mbps LAN is only about 20...

So whatever you play games on/watch HD TV over the internet on /watch
flash on, make that your primary OS.


I am not sure why one would want a virtual Linux machine tho. Except to
play with.


I had playing with computers in about 1984 when I started to program
them for a living..

Huh? IIRC Linux was not around in 1984.

What I want is a stable platform for the things I need to do, and thats
linux. I tolerate windows because I cat get two apps I need on any other OS.

I haven't noticed performance issues running VMs with VirtualBox. I
wouldn't even try running graphics intensive games though. OTOH MS
VirtualPC is sluggish for everything.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

"John Rumm" wrote in message
o.uk...

Because you were going on about Outlook which was not particularly helpful
in a discussion about Exchange.

I was never going on about exchange.
I never replied to anything about exchange.


Remember you asked "Since when did mail servers do POP to local files?",
so I told you when Exchange will do POP to local files.

You said some other add on to exchange now does POP.
However it doesn't do it to local files, it does it to the exchange database
if you want to get picky.
If you want it in your local file you need a client to get it and exchange
needs to be setup to let you have local files.


You can argue what you like but it doesn't change the facts.

I have no need to change the facts, I was just attempting to improve
your understanding of them.

I understand fully,

If that were the case, you would not have asked the question above...

I asked the question to get an answer, you did and it was half wrong.


you are arguing for the sake of it.

The irony...

It takes more than one to argue, so far you have been wrong about most of
it.

On Mon, 16 Apr 2012 17:32:28 +0100, John Rumm
wrote:

On 16/04/2012 15:36, Mark wrote:
On Mon, 16 Apr 2012 14:18:47 +0100, Clive George
wrote:

On 16/04/2012 13:58, Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.

It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc

Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?

Exists in C too, but in both languages you don't need to actually use it.

If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.

Part of learning is that sometimes things are provided which you
shouldn't use.

Yes. However I wouldn't first recommend making good practice a habit
by restricting the availability of "things you shouldn't use"

Does that not contradict what you said a few lines above about "...by
not even providing this feature", or am I missing something?

Well spotted ;-) The sentence contained a typo: s/wouldn't/would/

CF DIY and angle grinders...

I also would not recomend first starting DIY with an angle grinder.

Why not, if its the right tool for the job?

In this case I would recommend that a beginner get someone else to do
that job and start with something easier.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

On Mon, 16 Apr 2012 16:39:43 +0100, The Natural Philosopher
wrote:

Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.
It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc
Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?
Exists in C too, but in both languages you don't need to actually use it.

If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.


The point is that sometimes goto IS good practice.

Hmmmmm.

Its a fine way to make a simple clear direct statement for error
handling and stuff that cuts across - and has to cut across - the normal
flow of programming.

GOTO jail. Go directly to jail, Do NOT pass go. Do NOT collect £200.

That's Monopoly, not programming. If you were progamming a board game
then such a move could be done with a simple assignment statement
anyway.

Error handling can be done in a structured way using things like
exceptions or simple if/then constructs. Gotos make code hard to
read, maintain and test. In over 20 years of C programming I have
never needed to use a goto.

And "C" would also be a very poor language for teaching too.

Obviously anything that is pragmatically useful and written by engineers
to actually write real life code on is useless for academics who don't
want to sully their hands with the real world,

My point is to start with a language that (as much as possible)
enforces good practice. Once the fundamentals have been learned these
can be applied using other languages.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

On 16 Apr 2012 17:42:56 GMT, Bob Eager wrote:

On Mon, 16 Apr 2012 17:47:14 +0100, Tim Streater wrote:

In article ,
John Rumm wrote:

On 16/04/2012 16:39, The Natural Philosopher wrote:
Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.
It's OT for this thread, but for your information BBC BASIC is
currently
one of the most popular languages for teaching programming in UK
schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/
GCSE_Computing_J275_Centre_Crib_Sheet
_180112.doc

Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?
Exists in C too, but in both languages you don't need to actually
use it.

If the statement exists then the programmer can use it. In a
language used in teaching I would prefer the language would enforce
good practice by not even providing this feature.


The point is that sometimes goto IS good practice.

Its a fine way to make a simple clear direct statement for error
handling and stuff that cuts across - and has to cut across - the
normal flow of programming.

GOTO jail. Go directly to jail, Do NOT pass go. Do NOT collect £200.


And "C" would also be a very poor language for teaching too.

Obviously anything that is pragmatically useful and written by
engineers to actually write real life code on is useless for
academics who don't want to sully their hands with the real world,

Although you have to admit, that if said engineers had had decent
interactive terminals with screen editors etc, they may not have
insisted on something quit as concise and dependent on using every
printable character in the ASCII set for something! ;-)

Not sure about that. Remember, these are the ****ers that gave us TECO.
And two-character unix commands.

Nothing wrong with TECO! I met someone who had it solving differential
equations, too!

I once wrote an entire test system for a safety critical military
software project using editor macros because that's all I had (IIRC
This editor was far less sophisticated than TECO). It is an
experience I wouldn't want to repeat though.

And I still use an emacs-like editor.
--
(\__/) M.
(='.'=) If a man stands in a forest and no woman is around
(")_(") is he still wrong?

Mark wrote:
On Mon, 16 Apr 2012 22:00:48 +0100, The Natural Philosopher
wrote:

george [dicegeorge] wrote:
On 16/04/12 16:48, The Natural Philosopher wrote:
Mark wrote:
On Thu, 12 Apr 2012 12:38:29 +0100, John Rumm
wrote:

[-snip-]

However, XP mode is in reality a complete virtual machine running a
real copy of WinXP. There is nothing to stop you using any other
virtual PC hypervisor (including Microsoft's own Virtual PC) and
installing your own real copy of XP on that.
Running virtual machines can be very useful. However I would avoid
Microsoft Virtual PC. It's v-e-r-y slow. Oracle (Sun) VirtualBox is
much better IMHO.

Virtualbox is - adequate.

In order to get stable operation I have had to turn off MOST of the
'hardware acceleration' and that makes cursor response a but crappy, and
also flash videos don't really run smoothly under my XP machine.

But that's OK because of what I want my windows FOR. which is graphics
work.

Its useless for all but the simplest games, but its fine for RhinoCAD
and CorelDraw. And by doing what amounts to a suspend rather than a
shutdown, I can 'reboot' it in a few seconds.

For everything else there is Linux. Massively robust and functional.
Thouugh it did go doolally trying to transer a 15 Mbyte file via NFS to
a remote machines on the net...NFS is a fine protocol, but not over
links that occasioally do down..:-)




drat - virtualbox only 'adequate'?
I'm buying a new laptop,
my plan was to have Windows7 then virtual box then Xubuntu,
and I'd thought it would run almost as fast as if dualbooted.
Where is your slowdown coming from?

Video layer

simulating a video card + driver that maps to a windowed X windows is
not massively fast..

everything else is fine: in program computation gets pretty much as much
CPU as the basic linux timeslice allows, and its not a huge overhead in
context switching, so CPU intensive stuff is fine.

networking takes a BIT of a hit - again there is a 'virtual ethernet
card' presented.


But its for sure fast enough to access the internet full speed, even if
the 100Mbps LAN is only about 20...

So whatever you play games on/watch HD TV over the internet on /watch
flash on, make that your primary OS.


I am not sure why one would want a virtual Linux machine tho. Except to
play with.


I had playing with computers in about 1984 when I started to program
them for a living..

Huh? IIRC Linux was not around in 1984.

I never said it was.

I said that was when I stopped playing with computers.


What I want is a stable platform for the things I need to do, and thats
linux. I tolerate windows because I cat get two apps I need on any other OS.

I haven't noticed performance issues running VMs with VirtualBox. I
wouldn't even try running graphics intensive games though. OTOH MS
VirtualPC is sluggish for everything.


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

Mark wrote:
On Mon, 16 Apr 2012 16:39:43 +0100, The Natural Philosopher
wrote:

Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.
It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc
Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?
Exists in C too, but in both languages you don't need to actually use it.
If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.

The point is that sometimes goto IS good practice.

Hmmmmm.

Its a fine way to make a simple clear direct statement for error
handling and stuff that cuts across - and has to cut across - the normal
flow of programming.

GOTO jail. Go directly to jail, Do NOT pass go. Do NOT collect £200.

That's Monopoly, not programming. If you were progamming a board game
then such a move could be done with a simple assignment statement
anyway.

Error handling can be done in a structured way using things like
exceptions or simple if/then constructs. Gotos make code hard to
read, maintain and test. In over 20 years of C programming I have
never needed to use a goto.

And "C" would also be a very poor language for teaching too.
Obviously anything that is pragmatically useful and written by engineers
to actually write real life code on is useless for academics who don't
want to sully their hands with the real world,

My point is to start with a language that (as much as possible)
enforces good practice. Once the fundamentals have been learned these
can be applied using other languages.


My point is that the greatest error computer scientists made was to
believe that a computer language can enforce good practice, instead of a
proper educations and experienced programmers.

Its like having a language with no swear words. People will just kick
you instead.


--
To people who know nothing, anything is possible.
To people who know too much, it is a sad fact
that they know how little is really possible -
and how hard it is to achieve it.

Mark :
In over 20 years of C programming I have
never needed to use a goto.

My favourite C goto was in a published DefWindowProc. The author was
clearly embarrassed by the whole thing, so he used the destination label
ICantBelieveIActuallyUsedAGoto or something like that.

Of course he could have rewritten the code to avoid the goto, but
where's the fun in that?

--
Mike Barnes

"The Natural Philosopher" wrote in message
...
Mark wrote:
On Mon, 16 Apr 2012 16:39:43 +0100, The Natural Philosopher
wrote:

Mark wrote:
On Mon, 16 Apr 2012 12:31:23 +0100, Clive George
wrote:

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve

wrote:

Only ******* emote about BBC Basic, it's dead, Dave.
It's OT for this thread, but for your information BBC BASIC is
currently
one of the most popular languages for teaching programming in UK
schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc
Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?
Exists in C too, but in both languages you don't need to actually use
it.
If the statement exists then the programmer can use it. In a language
used in teaching I would prefer the language would enforce good
practice by not even providing this feature.

The point is that sometimes goto IS good practice.

Hmmmmm.

Its a fine way to make a simple clear direct statement for error
handling and stuff that cuts across - and has to cut across - the normal
flow of programming.

GOTO jail. Go directly to jail, Do NOT pass go. Do NOT collect £200.

That's Monopoly, not programming. If you were progamming a board game
then such a move could be done with a simple assignment statement
anyway.

Error handling can be done in a structured way using things like
exceptions or simple if/then constructs. Gotos make code hard to
read, maintain and test. In over 20 years of C programming I have
never needed to use a goto.

And "C" would also be a very poor language for teaching too.
Obviously anything that is pragmatically useful and written by engineers
to actually write real life code on is useless for academics who don't
want to sully their hands with the real world,

My point is to start with a language that (as much as possible)
enforces good practice. Once the fundamentals have been learned these
can be applied using other languages.

My point is that the greatest error computer scientists made was to
believe that a computer language can enforce good practice,

Corse it can if it has no gotos.

instead of a proper educations

Only fools need to be 'educated' in programming with a well designed
language.

and experienced programmers.

Its like having a language with no swear words.

Nothing like, actually.

People will just kick you instead.

Have fun listing how to do that with a computer language which has no gotos.

On Tue, 17 Apr 2012 11:06:59 +0100, Mike Barnes wrote:

Mark :
In over 20 years of C programming I have never needed to use a goto.

My favourite C goto was in a published DefWindowProc. The author was
clearly embarrassed by the whole thing, so he used the destination label
ICantBelieveIActuallyUsedAGoto or something like that.

Of course he could have rewritten the code to avoid the goto, but
where's the fun in that?

There was a language where a GOTO didn't necessarily go to where you
might expect:

label1:
...
label2:
...

then you could say:

label1 := label2

and then a goto to label1 would actually go to label2 ...!





--
Use the BIG mirror service in the UK:
http://www.mirrorservice.org

*lightning protection* - a w_tom conductor

On 16/04/2012 10:46, Mark wrote:
On Thu, 12 Apr 2012 17:34:34 +0100, "Richard Russell"
wrote:

On Wed, 11 Apr 2012 23:48:29 +0100, Steve
wrote:

Only ******* emote about BBC Basic, it's dead, Dave.

It's OT for this thread, but for your information BBC BASIC is currently
one of the most popular languages for teaching programming in UK schools.
For example it is recommended by the OCR examining board:

http://www.gcsecomputing.org.uk/support/index.html
http://social.ocr.org.uk/files/ocr/G...eet_180112.doc

Any flavour of BASIC is a poor choice of language for teaching
programming IMHO.

"GOTO" ... need I say more?
Bring back APL, I say. The only "write only" programming language I've used.

Pete