[DUG] What is the future for Delphi programmer?

[Jolyon Smith]

> I reckon you are still focusing way too narrowly.

Simply explained by the fact that I was responding to one very specific
observation, not offering a forward looking view covering all aspects of the
anticipated directions that Delphi may or may not ever go in.



> I am expecting office workers will still have desktops for years
> to come.  But more and more will be moving to phone and netbook
> devices, and already have started.

I might agree if I hadn't been hearing the exact same prediction for the
last 10 years.


> Windows desktops start to be replaced by terminal services
> and RDP devices to grunty servers.

Ditto.  This was "the way of the future" back in 1992.  I hasn't transpired
in the last 15 years.  What reason is there to think it will in the next 15?

I think the problem here is that there is a temptation to think that there
will be One True Way in the future, and we try to identify what that One
True Way will be.

Whereas the Truth is that there will always be "Horses For Courses".  There
may be a dominant approach, but even then what dominates in one area of
business/industry won't necessarily be appropriate for another, in which a
different approach will dominate.


> browsing, VNC, RDP connections to the servers.   (I have 
> been watching a colleague do all of these things already 
> from an Android phone... )

Then he's years behind.  I was doing this on a Windows Pocket PC back in
2000.

If the result of 9 years of technological progress in this area is simply
that now you can do exactly the same thing on some other handheld device,
then I struggle to see how this will lead to some new wave of computing
following this model.

If there was a need or a benefit, people would be doing it already.  This is
certain, because it has been possible.  If people aren't doing something
that they could be doing, there is usually a reason.

Of course, some people ARE doing it already and HAVE BEEN doing it for
years.  But that doesn't mean that EVERYONE will be doing it in the future.


> Phone was free on a 2 year Vodaphone plan.)

If you really think that that means the phone was "Free" then I have to say
that doesn't actually say much about your perception of reality.

;)


> Hardware keeps getting cheaper, instead of $500-$1500 for
> computers, it soon becomes $50-$500 for desktop RDP devices,
> netbooks and super-phones.

Again, this is the same mantra that has been churned out for the last 10-15
years.

Have we *really* already forgotten the NetPC ?

Again I ask the question... if the NetPC hasn't dominated the computing
world in the last 10 years, what reason do we have for thinking it will be
different in the next 10 years?

The technology isn't the issue.  Everything you describe has been possible
for years.  And I don't think price is the issue either.

On that score, your predictions of "cheap as chips" devices is naive
hardware companies will always find ways to justify maintaining higher price
tags.  Cheaper (older) technology then starts to get more expensive as
maintenance of that technology becomes more expensive due to its "legacy"
nature and the manufacturers seek to price it out of the market to compel
their customers to buy more new technology.

It was always thus and always will be.

The hardware companies simply aren't going to play your game I'm afraid.


And going back further of course, is simply an evolution of the mainframe
and dumb terminals architecture that the industry started out with, and with
"software as a service", goes full circle back to the days of purchasing
computer time.  I'm far too young to remember those days myself (I'm a
client/server baby), but I'm old enough to remember that such models
existed.

The users of the technology moved away from such models en mass, and
embraced a far less mature and in some ways less reliable and more expensive
model.

I suspect that the "vision of the future" we get these days is more a
reflection of the lack of awareness of what is in the rear-view mirror among
the "visionaries" in industry today, than it is based on any genuine
understanding of what business wants/needs from technology.


> I have seen examples of a linux device able to boot in 1-2
> seconds.

Whoop-de-doo.

My laptop takes 2 minutes or more to (cold) boot to a usable environment.

Reduce that to 7 seconds and you won't actually get any more productivity
out of me, because while it's booting I'm making myself my morning coffee.

:)


> I don't know the answer to that, but as long as phones 
> and netbooks keep outselling desktops the market dollars 
> will decide it.

Sales numbers aren't the end of the story.  At work I use a laptop and don't
have a desktop in the office at all.  But my laptop is used AS a desktop.
It get's docked first thing in the morning and stays locked in its dock all
day.

The purchase of that device contributed to the "sales of laptops
outstripping desktops" but the way the device is used means it should really
count in the desktop column.

Sales of laptops and mobile devices also reflect other aspects of those
devices...

When my laptop develops a fault in a USB port, GFX card or whatever, I have
to buy a new laptop.  Servicing a laptop rarely makes economic sense.

Kerching... "sales of laptops" gain another increment.


When the same thing happens on my desktop (at home), I simply replace the
defunct *part*.

When a faster CPU comes out/becomes cheap enough, I upgrade the CPU on my
desktop.

To get a faster laptop CPU I have to buy a new laptop.  Kerching.


It's fair to say I think that laptop sales should be adjusted to reflect
this obsolescence factor before being compared to desktop sales.

I don't know what that adjustment should be, but there should be one.



> More certainly, iPhone and Android devices will have a growing
> market, and I would love to be able to program for these without
> having to learn a new IDE and language....

I suggest that that attitude provides an insight into the genuine need you
perceive in these devices.

If you really have things that you can/want to do with those devices, then
you would not be waiting for Delphi to deliver to those platforms.

You would be learning what you needed to learn now, and employing those
devices in your solutions today?

No?