Ping those of a certain age.

could be worse.

 

hah - sorry - I was a bit pissed when I posted that. I wasn't
referring to "A." if that gives a slight clue though. Nothing more to
see here, move along, move along....
..

 

bad form, etc, but if you're not familiar with the song:

 

****, triple posting, but anyway:

"Lonely, alienated industrial man has unprecedented opportunities for
living life "in the context of equipment," as the philosopher Martin
Heidegger so aptly put it.

but...when I've used the phrase, I've meant it in the parodic form
used by the SF author Robert Sheckley in his book "Damocles"

 

Using the patented Mavis Beacon "Hunt&Peck" Technique, darsy

You are at serious risk of becoming an auto-parody here.

Just so you know...

 

Ah, you too eh? I think it's the sheer amount of spam hitting PC. Not
sure what happens next but SpamAssassin seems to get disabled as well.
I've had stuff not arriving for 12 hours or more, but I've also had some
emails go completely missing.

 

This is on top of the mailbox disappearing completely, and the website
being inaccessible for chunks of time. If my site is inaccessible I
actually use your site as well to see if it's just one server or their
whole outfit. And of course, these are just the ad-hoc times I need to
visit my own site. Goodness knows how often it's down when I've no idea.

If I could be bothered I'd set up a remote monitor.

I guess the forwarding is OK if the mail can get through to PC in the
first place.

Heh, Lightspeed Champion you ain't.

 

So now you're worse off no? If either services fails, you get no email.

Can't your android do IMAP direct to the PC server?

 

Or you could probably set it up so that all the champ.org stuff goes
direct to google, cutting out PC completely.

Fair enough. I'd like to meet the person who came up with the name cloud,
just to punch them in the face. I'm not sure why, but there it is.

 

I love the whole concept of putting business critical information and
applications where the internet has to be working to use them. Not.
It's all a bit back to the future.

 

I do something similar.
But when major corporations start doing it, that's another matter.

Jaguar is going all Cloudy in the Ford demerger at the mo, which will be
telling in whether it has matured.
www.silicon.com/retailandleisure/0,3800011842,39597173,00.htm

It could be a bit the same as the CTO deciding that Outsourcing is a
wonderful idea. Primarily because he/she has failed in a number of
respects, particularly in strategic decisions and
recruiting/retaining/motivating technical staff. Or indeed for not
*actually* knowing anything technical at all.

I'm sure you remember as well as me why people went to distributed
servers and client applications. But yeah the world of hardware and
fibre has moved on some.

 

From what I've read Google's entire app suite is fully distributed.
The data always exists in several datacentres at the same time, and
they do "invisible" load balancing in front of that.

There's an unsubstantiated rumour that the outages they've had occured
when completely failing over entire datacentres hasn't worked as
perfectly as it should, and if that's the case I expect it's something
they'll get re-written in fairly short order.

 

Oh they are but how many big internet tribs do companies have, thanks to the
fibre policies of BT? And if they have redundant connections do they go to
different BT exchanges? That old BT trick of damned expensive Diverse Fibre
Routing.... to the same Exchange service. That's something they were always
quite prepared to lie about.

I meant distributed as in having a server on the LAN segment for the users
on that segment.

When I designed the EPOS and sales intranet for DX the local part of the
central database and apps was replicated out to a small server in each
retail outlet. When the central services went down or a frame relay node
lost connectivity it was business as usual bar phoning for credit vets. They
worked on the local server which replicated data up, if the local service
failed they worked on the central service. Sure it was expensive but it paid
for itself in so many ways a hundred times over.

Carphone Whorehouse have been cursed with a centralised system, lets call it
a Cloud system! since the same time. It has driven sales staff to suicide.
But Dunstan "bought the company" and it has survived to this day. I bought a
phone and modem in CPW last November and it was comical the time it took,
with performance lag on the system (busy Sat) and then it all fell over and
the shop stopped trading for 10 mins while central services were resumed.

 

I'd prefer to still call it a centralised system.

My take on cloud is that it's more often than not decentralised with
apps and processes capable of running on more than one site, and
easily switch on/off-able. The whole concept about cloud is that the
infrastructure isn't rigid.

I do agree with you that totally centralising systems by removing all
local resiliance is a backwards step though. Not that it stops
companies wanting to do it.

 

Yes I agree completely and it's pretty what I was getting at. And it is
driven in part by a lack of technical expertise and part of that is CTO's
and IT Directors with no real IT ability or knowledge. But MD's, CEO's and
CFO's have to carry the can for putting the wrong people in the frame. It's
like selecting a doctor to operate on you who did a medical degree then went
straight to being a medical sales rep.

 

I think it harks back to when solutions architects all used to draw
pretty pictures when the internet was a fluffy bubble.

 

heh, I've got recent experience of two large banks in this country
hosting on Amazon's cloud...

 

It still is a fluffy bubble, and long may it's fluffy bubblyness
continue.

 

who doesn't?

What other shape would "the internet" be?

 

Yes, got the one I wanted!