I've read scuttlebutt about M$ moving Windows to a SaaS model, but I fail to
understand how this could possibly work.

Leaving aside what I personally think of the whole Software as a Service model,
I can understand, after a fashion, why it might be appealing for a development
team to be able to issue mass updates to subscribers, rather than waiting for
people to upgrade and continuing to offer backwards support and compatibility
for legacy versions.

And the idea of having a few.. hundred... million customers paying you $20 a
month forever sure must seem like a Good Idea to the executives of companies
that already have more money than Croesus.

That said: I personally believe that it would stifle innovation, since an
independent developer couldn't test a program on their own machine, but would be
forced to run it in a sandbox VM.

This is pretty much the dilemma I'm having with iOS apps at the moment.  I
despise what Apple has done to iTunes, and there are features that I would love
to implement into a music player, but unless I jailbreak my iPod, I can't write
a program solely for my own, personal edification, and use it.  In order to get
the app on my iPod, as I understand it, I would have to have it digitally
signed, for which I would have to pay, and have the app reviewed by Apple (which
requires showing them the code -- yeah... No). Or have an Enterprise Developer
license which costs an insane amount of money.

In reality, I understand that there are probably music players already on the
app store that do most, if not all, of the things I want to do; however, sifting
through them is a pain, and I could potentially learn something by doing it
myself.

But I digress.

As I aspire to, one day, program for a living -- A topic not for discussion on a
public forum -- I am interested in the various development and marketing models
used in the industry, and often find myself at odds with various philosophies,
but I try to remain open minded.

Anybody else have thoughts on the SaaS model?

Regards,
A.D.B.

P.S. If anyone knows of a good music player for the iOS that sports an
attractive interface and allows you to construct and save ad-hoc playlists on
the fly (e.g. building a temporary playlist while playing the songs on it.) I
would appreciate suggestions.

Post a reply to this message

On Fri, 31 Jul 2015 00:12:26 -0400, Anthony D. Baye wrote:

> Anybody else have thoughts on the SaaS model?

Having worked for both a traditional software company and a software 
company that currently offers both SaaS and on-premise solutions, I can 
see why the SaaS model is attractive to a software company.

Recurring revenue helps you grow the business.

Traditional perpetual license models mean that the customer buys the 
product once.  They may have to license it for a number of users or 
servers, but ultimately, the cost is pretty low over the lifetime of the 
product.

That's good for the customer that doesn't want to continually upgrade.

Most customers, though, want the latest and greatest features and 
functionality - and a subscription model (which is common with SaaS 
products), done properly, funds ongoing development.  Anyone who has 
worked retail knows that the cost of acquiring a new customer is pretty 
high; the cost of retaining a customer is relatively low, as long as you 
keep the customer happy.  The SaaS subscription model works off that 
fundamental idea in retail business.

In order to retain customers, you have to listen to them and implement 
what they're looking for.  That's a lot easier when you can afford to pay 
developers a decent salary and can grow the development team (and design 
team - design is very important in software development, and a lot of 
companies fail to design well before implementing the product).

If you'd asked me 10 years ago what I thought of SaaS, I'd have said that 
I thought it was a scam.  Now that I've seen it up close in a company 
that sells with that model, and have had the growth trajectory explained 
by financial people who understand it *and* who understand how to explain 
it, I see the benefits, both to the company and to the customer and end 
user.

A loyal customer for a software company isn't a customer who buys once 
and never comes back - even if they never change products.  A loyal 
customer is a customer who continues to fund further development so the 
developers can continue to improve the product for the benefit of the 
customer.

Jim

-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On 7/30/2015 9:48 PM, Jim Henderson wrote:
> If you'd asked me 10 years ago what I thought of SaaS, I'd have said that
> I thought it was a scam.  Now that I've seen it up close in a company
> that sells with that model, and have had the growth trajectory explained
> by financial people who understand it *and* who understand how to explain
> it, I see the benefits, both to the company and to the customer and end
> user.
>
> A loyal customer for a software company isn't a customer who buys once
> and never comes back - even if they never change products.  A loyal
> customer is a customer who continues to fund further development so the
> developers can continue to improve the product for the benefit of the
> customer.
>
> Jim
>
However... With respect, I already pay like $30-40 a month for access to 
certain online games. To suggest that I have to pay for the OS they run 
on as well, and a similar amount.. is, to me, a huge fuck you to me, and 
anyone either unwilling, or unable, to keep paying for such a thing. 
So.. No, I can see a "service", or even some "software", being paid for 
this way, though I despise the idea for something I need only rarely, 
but... to pay for an OS in such a manner is tantamount to making someone 
rent the car, then the tires, then the gas in the tank, then have a 
special license to drive on your own street, another to drive in the 
city, and yet another to drive highways, and on, and on, and on.

It may be a wet dream, to the assholes doing it, but, its precisely the 
reason everyone is against the attempts to kill net neutrality, among 
other one sided, highway robbery, style attempts to make every single 
fart cost you money, before you even **get** to the actual feature you 
are trying to use in the first place.

So, no... its still a scam, when used to bludgeon people into paying for 
things they need, but can't otherwise get. And, if Microsoft thinks this 
is a sensible method... then maybe someone should ask them why, then, 
they "sell" the xbox, instead of just renting it, for one example.... It 
amounts to the same thing.

-- 
Commander Vimes: "You take a bunch of people who don't seem any 
different from you and me, but when you add them all together you get 
this sort of huge raving maniac with national borders and an anthem."

Post a reply to this message

> I've read scuttlebutt about M$ moving Windows to a SaaS model, but I fail to
> understand how this could possibly work.

Locally you'd have an OS that was stripped down to just run Remote 
Desktop (or equivalent) and interface with your hardware. When you 
logged on it would start a remote desktop session with an MS VM 
somewhere. *Assuming internet speeds were fast enough* you wouldn't 
notice the difference to running full windows locally.

The benefits are obvious (a machine that has all your files and looks 
the same no matter where you log on, an almost limitless supply of CPU 
power and RAM if you wanted to do CPU intenstive tasks, automatic 
backups for everyone, etc), I fail to understand how you fail to 
understand how this could possibly work :-)

> Leaving aside what I personally think of the whole Software as a Service model,
> I can understand, after a fashion, why it might be appealing for a development
> team to be able to issue mass updates to subscribers, rather than waiting for
> people to upgrade and continuing to offer backwards support and compatibility
> for legacy versions.

Their big problem will be the medium-large corporations that take 
months, if not years to test and roll out major software updates. There 
is no way they would accept the possibility of one day their entire 
company coming to a halt with millions of pounds lost due to an MS 
"update" that has broken something somewhere within their business. Also 
a lot of systems are not connected to the internet for various reasons, 
how would they work?

> That said: I personally believe that it would stifle innovation, since an
> independent developer couldn't test a program on their own machine, but would be
> forced to run it in a sandbox VM.

If the whole of windows and all apps are running on a remote VM, then I 
don't see why you wouldn't just fire up Visual Studio (or whatever) on 
the same remote machine. You'd be using exactly the same environment as 
everyone else to develop and test your app.

> This is pretty much the dilemma I'm having with iOS apps at the moment.  I
> despise what Apple has done to iTunes, and there are features that I would love
> to implement into a music player, but unless I jailbreak my iPod, I can't write
> a program solely for my own, personal edification, and use it.

I see no indication that MS would follow such a route, if they wanted to 
they could have done so already. That's just Apple being Apple, and 
their philosophy that if they control everything the user will have a 
better experience.

> In reality, I understand that there are probably music players already on the
> app store that do most, if not all, of the things I want to do; however, sifting
> through them is a pain, and I could potentially learn something by doing it
> myself.

Sounds to me like you should have bought an Android phone. One click in 
Eclipse (free) on my desktop PC and my app is compiled, sent to my phone 
and running within a few seconds.

> As I aspire to, one day, program for a living -- A topic not for discussion on a
> public forum -- I am interested in the various development and marketing models
> used in the industry, and often find myself at odds with various philosophies,
> but I try to remain open minded.

If you're interested you might like to try and do some financial 
modelling of the SaaS model for Windows. You can probably find out how 
many copies of Windows MS sells, how much income they've got from that 
over the years etc. Try and predict that if they continued with the same 
model how well would the company be doing in 5 years, 10 years, 20 years 
time.

Then assume they switched to a SaaS model at some point in the future. 
You'd need to make many assumptions about many different items. But you 
should be able to predict the profit they would make after such a 
switch. How does it compare to your previous predictions? How sensitive 
is it to changes in your assumptions? What is the optimum subscription 
price? How to handle business users? When it the optimum time to stop 
support for old versions of Windows?

No doubt MS has teams of people doing exactly the above (with access to 
far more information than publically available). *If* the calculations 
show more profit with low enough risk can be achieved with the SaaS 
model then the executives will make the decision.

Post a reply to this message

On 7/31/2015 9:56 AM, scott wrote:
>> I've read scuttlebutt about M$ moving Windows to a SaaS model, but I
>> fail to
>> understand how this could possibly work.
>
> Locally you'd have an OS that was stripped down to just run Remote
> Desktop (or equivalent) and interface with your hardware. When you
> logged on it would start a remote desktop session with an MS VM
> somewhere. *Assuming internet speeds were fast enough* you wouldn't
> notice the difference to running full windows locally.
>
> The benefits are obvious (a machine that has all your files and looks
> the same no matter where you log on, an almost limitless supply of CPU
> power and RAM if you wanted to do CPU intenstive tasks, automatic
> backups for everyone, etc), I fail to understand how you fail to
> understand how this could possibly work :-)

No, no, no!

Have you ever worked on a thin client?
I'd rather use a Teletype as a dumb terminal.

;)

-- 

Regards
     Stephen

Post a reply to this message

> No, no, no!
>
> Have you ever worked on a thin client?
> I'd rather use a Teletype as a dumb terminal.
>
> ;)

You need a fatter client and/or a fatter pipe then :-)

Post a reply to this message

On 7/31/2015 2:02 PM, scott wrote:
>> No, no, no!
>>
>> Have you ever worked on a thin client?
>> I'd rather use a Teletype as a dumb terminal.
>>
>> ;)
>
> You need a fatter client and/or a fatter pipe then :-)
>

On my last contract, even the local IT manager hated it.

-- 

Regards
     Stephen

Post a reply to this message

>>> Have you ever worked on a thin client?
>>> I'd rather use a Teletype as a dumb terminal.
>>>
>>> ;)
>>
>> You need a fatter client and/or a fatter pipe then :-)
>>
>
> On my last contract, even the local IT manager hated it.

I don't know what you mean by "it" - here we use remote desktop pretty 
much daily to work on machines in the labs (mainly because we are lazy 
and don't want to get up). Once you make it full screen it's easy to 
forget you're actually in remote desktop rather than your native OS.

However I've used it from home a few times and it's more obvious you're 
on remote desktop, window dragging is not as smooth, updates are 
noticeably slower. Hence my comment that once internet speeds are fast 
enough I see those issues going away for RD.

But from the other angle you have things like Office and Google Docs 
that work in a browser, and using those even on relatively slow 
connections seem fine. So maybe Windows SaaS could be no more than a 
"start page" in a browser used to launch other web-based apps? Who 
knows, there are many possibilities.

Post a reply to this message

On 7/31/2015 2:49 PM, scott wrote:

> I don't know what you mean by "it" -

Information Technology, young man.

> here we use remote desktop pretty
> much daily to work on machines in the labs (mainly because we are lazy
> and don't want to get up). Once you make it full screen it's easy to
> forget you're actually in remote desktop rather than your native OS.
>
> However I've used it from home a few times and it's more obvious you're
> on remote desktop, window dragging is not as smooth, updates are
> noticeably slower. Hence my comment that once internet speeds are fast
> enough I see those issues going away for RD.
>

I think that you are in a privileged position. You work in the 
automotive industry and in R&D. When I was working at Mercedes the IT 
systems were top notch. Really good, you wouldn't know that you were 
working remotely. Unfortunately most of the companies and sites I've 
been at that is not the case. One site I worked at we called the system 
Godot as we were always waiting for it. ;-)
The number of times I have forgotten what I was going to do by the time 
the machine had drawn the screen are legend. Not all the time of course 
but enough to raise raise your frustration levels to boiling point.


> But from the other angle you have things like Office and Google Docs
> that work in a browser, and using those even on relatively slow
> connections seem fine. So maybe Windows SaaS could be no more than a
> "start page" in a browser used to launch other web-based apps? Who
> knows, there are many possibilities.
>

I wish I lived in your world. :)

-- 

Regards
     Stephen

Post a reply to this message

> I think that you are in a privileged position. You work in the
> automotive industry and in R&D.

You must have a good memory, because I quit that job over 3 years ago 
:-) This place makes industrial printing equipment. Still I think any 
bargain PC can handle remote desktop, the bottleneck seems to be the 
internet connection between my house network and my work network. Remote 
desktop works seamlessly within each network, but try to access one from 
the other and it is very awkward to use.

> The number of times I have forgotten what I was going to do by the time
> the machine had drawn the screen are legend. Not all the time of course
> but enough to raise raise your frustration levels to boiling point.

Oh that happens regularly here too on other system, just not seen it on 
remote desktop across the same network. Try getting in to work and then 
waiting over 30 mins for your PC to boot. I just lock mine over night now.

> I wish I lived in your world. :)

I'd find out a bit more about it first if I were you :-)

Post a reply to this message

On Thu, 30 Jul 2015 22:49:00 -0700, Patrick Elliott wrote:

> On 7/30/2015 9:48 PM, Jim Henderson wrote:
>> If you'd asked me 10 years ago what I thought of SaaS, I'd have said
>> that I thought it was a scam.  Now that I've seen it up close in a
>> company that sells with that model, and have had the growth trajectory
>> explained by financial people who understand it *and* who understand
>> how to explain it, I see the benefits, both to the company and to the
>> customer and end user.
>>
>> A loyal customer for a software company isn't a customer who buys once
>> and never comes back - even if they never change products.  A loyal
>> customer is a customer who continues to fund further development so the
>> developers can continue to improve the product for the benefit of the
>> customer.
>>
>> Jim
>>
> However... With respect, I already pay like $30-40 a month for access to
> certain online games. To suggest that I have to pay for the OS they run
> on as well, and a similar amount.. 

Not all operating systems are free.  Someone has to develop them, and 
those developers need to eat.

But my comment was more about SaaS in general, not OSaaS.

Jim
-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On 7/31/2015 4:03 PM, scott wrote:
>> I think that you are in a privileged position. You work in the
>> automotive industry and in R&D.
>
> You must have a good memory, because I quit that job over 3 years ago
> :-) This place makes industrial printing equipment.

My wife might not agree with you. :)
I think that you have mentioned the industry since you left.

> Still I think any
> bargain PC can handle remote desktop, the bottleneck seems to be the
> internet connection between my house network and my work network. Remote
> desktop works seamlessly within each network, but try to access one from
> the other and it is very awkward to use.
>

Probably just the crappy companies I've worked for.

>> The number of times I have forgotten what I was going to do by the time
>> the machine had drawn the screen are legend. Not all the time of course
>> but enough to raise raise your frustration levels to boiling point.
>
> Oh that happens regularly here too on other system, just not seen it on
> remote desktop across the same network. Try getting in to work and then
> waiting over 30 mins for your PC to boot. I just lock mine over night now.
>

Routine:
Arrive, switch on PC, take of jacket, go for coffee, log on, talk to 
early starters. Look to see if it is ready. :)

>> I wish I lived in your world. :)
>
> I'd find out a bit more about it first if I were you :-)

I know. The other man's grass is always stronger. ;-)

-- 

Regards
     Stephen

Post a reply to this message

On 31/07/2015 09:56 AM, scott wrote:
>> I've read scuttlebutt about M$ moving Windows to a SaaS model, but I
>> fail to
>> understand how this could possibly work.
>
> Locally you'd have an OS that was stripped down to just run Remote
> Desktop (or equivalent) and interface with your hardware. When you
> logged on it would start a remote desktop session with an MS VM
> somewhere. *Assuming internet speeds were fast enough* you wouldn't
> notice the difference to running full windows locally.

The more I look at the IT world, the more cyclic it seems to be.

There was a time when you bought the biggest, most powerful mainframe 
that money could buy, and all the users sat at dumb terminals logged in 
to the giant monster in the middle.

And then everybody said "hey, putting a single desktop at each person's 
desk means you can more easily add and remove computer power depending 
on staffing levels, equip different people with different versions of 
software, etc."

And now everybody's like "hey, it's a pain to manage multiple individual 
desktops. Let's virtualise everything to get a bigger return on 
investment..."

And so the industry continues to alternate between centralised and 
decentralised. Because, frankly, each has different pros and cons; it's 
just that every decade or so people forget the pros of one and forget 
the cons of the other.

> The benefits are obvious (a machine that has all your files and looks
> the same no matter where you log on, an almost limitless supply of CPU
> power and RAM if you wanted to do CPU intenstive tasks, automatic
> backups for everyone, etc)

I think you mean "we can give you less and less CPU and RAM while still 
charging the same amount of money for it, so you will continually have 
to give us more money or suffer horrendously unusable system response".

And then of course, you have the problem that each morning, you log into 
your desktop, and there's a 50% probability that the software will have 
changed, and you can't prevent it changing. Already we see every time 
Facebook changes the colour of a button, somebody creates a page 
entitled "if one million users Like this page, Facebook will turn the 
button colour back to how it was before". [Erm, no they won't honey.] 
Imagine if every day, all your software could be deleted and replaced 
with something else that you didn't ask for or want.

To say nothing of the privacy and confidentiality issues of having 
Microsoft have access to every file you ever create. (I doubt too many 
corporate types would like having their propriety data on a hostile 3rd 
party server.)

> Their big problem will be the medium-large corporations that take
> months, if not years to test and roll out major software updates. There
> is no way they would accept the possibility of one day their entire
> company coming to a halt with millions of pounds lost due to an MS
> "update" that has broken something somewhere within their business. Also
> a lot of systems are not connected to the internet for various reasons,
> how would they work?

They also have a problem with SOHO setups where people wouldn't know 
what "computer security" is if it hit them in the face.

Why no, I'm not bitter. Why do you ask?

Post a reply to this message

On 31/07/2015 05:48 AM, Jim Henderson wrote:
> Recurring revenue helps you grow the business.
>
> Traditional perpetual license models mean that the customer buys the
> product once.  They may have to license it for a number of users or
> servers, but ultimately, the cost is pretty low over the lifetime of the
> product.
>
> That's good for the customer that doesn't want to continually upgrade.
>
> Most customers, though, want the latest and greatest features and
> functionality - and a subscription model (which is common with SaaS
> products), done properly, funds ongoing development.  Anyone who has
> worked retail knows that the cost of acquiring a new customer is pretty
> high; the cost of retaining a customer is relatively low, as long as you
> keep the customer happy.  The SaaS subscription model works off that
> fundamental idea in retail business.

Selling "support" also works. Provided you have the staff to actually 
deliver support.

Or just a licence to download any and all new versions that get put out 
in the next X months. Keep paying if you want new stuff.

Or paying to have a customised version of the product. (But that's sort 
of tantamount to saying "our product is so hard to set up, you need to 
pay us to do it for you". Not a great thing to say!)

Or hell, even paying for the development of specific features that you 
particularly need. The trouble with this one is that

* Usually there's only one or two specific features a given customer 
needs. Once those are done, no revenue stream.

* Usually how this *actually* plays out is the customer says "we won't 
buy it until it does X". And then you implement X. And then they still 
don't buy. (Why no, I'm not bitter. Why do you keep asking that?)

Certainly where I currently work, a lot of customers specifically *like* 
the fact that if they stop paying, the software doesn't self-destruct. 
Nobody is going to lose any data. You can carry on using the software 
forever; you just won't get any bugfixes or new features.

Then again, where I work, it takes most customers up to 3 years to get 
authorisation to buy a new mouse. So convincing people to buy at all is 
very hard.

(If the second-hand stuff I've heard is at all accurate, the people who 
will *use* the software utterly love it and really want to buy it - but 
the people who authorise the money won't let you buy anything without 
having it signed in the blood of a unicorn...)

Post a reply to this message

On Fri, 31 Jul 2015 19:37:24 +0100, Orchid Win7 v1 wrote:

> On 31/07/2015 09:56 AM, scott wrote:
>>> I've read scuttlebutt about M$ moving Windows to a SaaS model, but I
>>> fail to understand how this could possibly work.
>>
>> Locally you'd have an OS that was stripped down to just run Remote
>> Desktop (or equivalent) and interface with your hardware. When you
>> logged on it would start a remote desktop session with an MS VM
>> somewhere. *Assuming internet speeds were fast enough* you wouldn't
>> notice the difference to running full windows locally.
> 
> The more I look at the IT world, the more cyclic it seems to be.

Yep, I've observed this for over 30 years myself.

>> The benefits are obvious (a machine that has all your files and looks
>> the same no matter where you log on, an almost limitless supply of CPU
>> power and RAM if you wanted to do CPU intenstive tasks, automatic
>> backups for everyone, etc)
> 
> I think you mean "we can give you less and less CPU and RAM while still
> charging the same amount of money for it, so you will continually have
> to give us more money or suffer horrendously unusable system response".
> 
> And then of course, you have the problem that each morning, you log into
> your desktop, and there's a 50% probability that the software will have
> changed, and you can't prevent it changing. Already we see every time
> Facebook changes the colour of a button, somebody creates a page
> entitled "if one million users Like this page, Facebook will turn the
> button colour back to how it was before". [Erm, no they won't honey.]
> Imagine if every day, all your software could be deleted and replaced
> with something else that you didn't ask for or want.
> 
> To say nothing of the privacy and confidentiality issues of having
> Microsoft have access to every file you ever create. (I doubt too many
> corporate types would like having their propriety data on a hostile 3rd
> party server.)

Surprisingly, a lot of companies are moving to SaaS solutions for a lot 
of things.  I'm constantly amazed at the new customers my employer brings 
in to help secure their SaaS infrastructures using identity as the 
"border".

One of the larger challenges in modern IT is that the old paradigm of 
securing everything with a firewall is done.  The internal network isn't 
a trusted network these days, even if you use no SaaS solutions.  Look at 
Sony.  Look at Target.  Look at any major company that has been hacked in 
the past 5 years - all of them trusted their internal networks, and it 
cost them huge.

The "firewall" concept is flawed, and these days, proper security is 
identity-enabled and tied to devices and users rather than a network.

The other surprising thing is that as the "Internet of Things" takes off, 
these "things" are not being designed with security in mind - security is 
being bolted on after the fact.

>> Their big problem will be the medium-large corporations that take
>> months, if not years to test and roll out major software updates. There
>> is no way they would accept the possibility of one day their entire
>> company coming to a halt with millions of pounds lost due to an MS
>> "update" that has broken something somewhere within their business.
>> Also a lot of systems are not connected to the internet for various
>> reasons, how would they work?
> 
> They also have a problem with SOHO setups where people wouldn't know
> what "computer security" is if it hit them in the face.
> 
> Why no, I'm not bitter. Why do you ask?

Actually, the IDaaS market in the SME space is going to be a very bloody 
battle between Microsoft and Google.  There's really no reason for anyone 
else to even play in that market.  Larger enterprises will tend to look 
elsewhere (though there are a number of > 10,000 user companies that use 
Google, apparently).

Jim
-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On 7/31/2015 7:37 PM, Orchid Win7 v1 wrote:
> And then of course, you have the problem that each morning, you log into
> your desktop, and there's a 50% probability that the software will have
> changed, and you can't prevent it changing.

And if you want to use something non standard. It is a PITA to get 
installed.
I used to use a small graphics tablet instead of a mouse. In the end I 
gave up using it because of the hassle getting the drivers installed.
And if you wanted to use the machine to run PovRay overnight or at 
lunchtime...

-- 

Regards
     Stephen

Post a reply to this message

On 7/31/2015 7:47 PM, Orchid Win7 v1 wrote:
>
> Selling "support" also works. Provided you have the staff to actually
> deliver support.
>

The sector I work in, often supply the software customisation of the 
s/ware at what they hope is break even prices. To get the support contract.

> Or just a licence to download any and all new versions that get put out
> in the next X months. Keep paying if you want new stuff.
>
> Or paying to have a customised version of the product. (But that's sort
> of tantamount to saying "our product is so hard to set up, you need to
> pay us to do it for you". Not a great thing to say!)
>

Shhh!

> Or hell, even paying for the development of specific features that you
> particularly need. The trouble with this one is that
>
> * Usually there's only one or two specific features a given customer
> needs. Once those are done, no revenue stream.

Not necessarily. SAP helped Shell configure their system for offshore 
logistics. Then sold the solution to other oil companies as an Industry 
Solution. They did the same with Rolls Royce Aero-systems and created an 
expensive IS-Aero module. Which they sell to aircraft companies. (That 
is the system that will be able to tell the investigators of the MH370 
crash. Which aeroplane the flaperon came from.)

-- 

Regards
     Stephen

Post a reply to this message

On 7/31/2015 7:47 PM, Orchid Win7 v1 wrote:
 >
 > Selling "support" also works. Provided you have the staff to actually
 > deliver support.
 >

The sector I work in, often supply the software customisation at what 
they hope is break even prices. To get the support contract.

 > Or just a licence to download any and all new versions that get put out
 > in the next X months. Keep paying if you want new stuff.
 >
 > Or paying to have a customised version of the product. (But that's sort
 > of tantamount to saying "our product is so hard to set up, you need to
 > pay us to do it for you". Not a great thing to say!)
 >

Shhh!

 > Or hell, even paying for the development of specific features that you
 > particularly need. The trouble with this one is that
 >
 > * Usually there's only one or two specific features a given customer
 > needs. Once those are done, no revenue stream.

Not necessarily. SAP helped Shell configure their system for offshore 
logistics. Then sold the solution to other oil companies as an Industry 
Solution. They did the same with Rolls Royce Aero-systems and created an 
expensive IS-Aero module. Which they sell to aircraft companies. (That 
is the system that will be able to tell the investigators of the MH370 
crash. Which aeroplane the flaperon came from.)

-- 

Regards
     Stephen

Post a reply to this message

Orchid Win7 v1 <voi### [at] devnull> wrote:
> On 31/07/2015 09:56 AM, scott wrote:
> >> I've read scuttlebutt about M$ moving Windows to a SaaS model, but I
> >> fail to
> >> understand how this could possibly work.
> >
> > Locally you'd have an OS that was stripped down to just run Remote
> > Desktop (or equivalent) and interface with your hardware. When you
> > logged on it would start a remote desktop session with an MS VM
> > somewhere. *Assuming internet speeds were fast enough* you wouldn't
> > notice the difference to running full windows locally.
>
> The more I look at the IT world, the more cyclic it seems to be.
>
> There was a time when you bought the biggest, most powerful mainframe
> that money could buy, and all the users sat at dumb terminals logged in
> to the giant monster in the middle.
>
> And then everybody said "hey, putting a single desktop at each person's
> desk means you can more easily add and remove computer power depending
> on staffing levels, equip different people with different versions of
> software, etc."
>
> And now everybody's like "hey, it's a pain to manage multiple individual
> desktops. Let's virtualise everything to get a bigger return on
> investment..."
>
> And so the industry continues to alternate between centralised and
> decentralised. Because, frankly, each has different pros and cons; it's
> just that every decade or so people forget the pros of one and forget
> the cons of the other.
>
> > The benefits are obvious (a machine that has all your files and looks
> > the same no matter where you log on, an almost limitless supply of CPU
> > power and RAM if you wanted to do CPU intenstive tasks, automatic
> > backups for everyone, etc)
>
> I think you mean "we can give you less and less CPU and RAM while still
> charging the same amount of money for it, so you will continually have
> to give us more money or suffer horrendously unusable system response".
>
> And then of course, you have the problem that each morning, you log into
> your desktop, and there's a 50% probability that the software will have
> changed, and you can't prevent it changing. Already we see every time
> Facebook changes the colour of a button, somebody creates a page
> entitled "if one million users Like this page, Facebook will turn the
> button colour back to how it was before". [Erm, no they won't honey.]
> Imagine if every day, all your software could be deleted and replaced
> with something else that you didn't ask for or want.
>
> To say nothing of the privacy and confidentiality issues of having
> Microsoft have access to every file you ever create. (I doubt too many
> corporate types would like having their propriety data on a hostile 3rd
> party server.)
>
> > Their big problem will be the medium-large corporations that take
> > months, if not years to test and roll out major software updates. There
> > is no way they would accept the possibility of one day their entire
> > company coming to a halt with millions of pounds lost due to an MS
> > "update" that has broken something somewhere within their business. Also
> > a lot of systems are not connected to the internet for various reasons,
> > how would they work?
>
> They also have a problem with SOHO setups where people wouldn't know
> what "computer security" is if it hit them in the face.
>
> Why no, I'm not bitter. Why do you ask?

The difference between a company with a gigantic mainframe that all the
employees log into from a single campus and the idea of Windows as a Service, is
a pretty big one, IMHO.

Now we have people whose office is the world.  They take their computers into
the field where they sometimes won't have internet access for days on end, and
when they do, it's in some third world country over a satellite feed that has
limited bandwidth.

Then too, at what point do you think that the governments of the world will
become comfortable with their internal communications being stored -- even
temporarily -- in the buffer of someone else's computers? Personally, I think
they'll go back to using typewriters first, but realistically, they'll probably
just switch to linux.

Regards,
A.D.B.

Post a reply to this message

On Fri, 31 Jul 2015 19:47:14 +0100, Orchid Win7 v1 wrote:

> Selling "support" also works. Provided you have the staff to actually
> deliver support.

The problem with this is that if you're selling support, your product is 
poorly designed and difficult to use.

If you're a software company that depends on support and training as 
bottom-line revenue figures, you're doing software wrong IMHO.

Jim

-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On 01/08/2015 12:09 AM, Jim Henderson wrote:
> On Fri, 31 Jul 2015 19:47:14 +0100, Orchid Win7 v1 wrote:
>
>> Selling "support" also works. Provided you have the staff to actually
>> deliver support.
>
> The problem with this is that if you're selling support, your product is
> poorly designed and difficult to use.
>
> If you're a software company that depends on support and training as
> bottom-line revenue figures, you're doing software wrong IMHO.

Heh. Well there is that... ;-)

Post a reply to this message

Le 2015-07-31 19:09, Jim Henderson a écrit :
> On Fri, 31 Jul 2015 19:47:14 +0100, Orchid Win7 v1 wrote:
>
>> Selling "support" also works. Provided you have the staff to actually
>> deliver support.
>
> The problem with this is that if you're selling support, your product is
> poorly designed and difficult to use.
>
> If you're a software company that depends on support and training as
> bottom-line revenue figures, you're doing software wrong IMHO.
>

You're saying a oil-rig management system, or an MRI control system 
shouldn't require training?

I agree that if you're selling 3D-tictactoe-as-a-Service, and expect 
people to call you for support, you may have a problem, but complex 
software such as ERPs, or even industry-grade CAD systems _should_ 
require some level of training to install and operate properly.

> Jim
>


-- 
/*Francois Labreque*/#local a=x+y;#local b=x+a;#local c=a+b;#macro P(F//
/*    flabreque    */L)polygon{5,F,F+z,L+z,L,F pigment{rgb 9}}#end union
/*        @        */{P(0,a)P(a,b)P(b,c)P(2*a,2*b)P(2*b,b+c)P(b+c,<2,3>)
/*   gmail.com     */}camera{orthographic location<6,1.25,-6>look_at a }

Post a reply to this message

Am 01.08.2015 um 01:09 schrieb Jim Henderson:
> On Fri, 31 Jul 2015 19:47:14 +0100, Orchid Win7 v1 wrote:
>
>> Selling "support" also works. Provided you have the staff to actually
>> deliver support.
>
> The problem with this is that if you're selling support, your product is
> poorly designed and difficult to use.
>
> If you're a software company that depends on support and training as
> bottom-line revenue figures, you're doing software wrong IMHO.

Support := help with operating the software in conditions not 
anticipated during the design.

Nothing wrong with that in my book, provided the company isn't shifting 
costs from the design phase to the support phase.

Sample case: A big company I worked for some day realized that they were 
growing out of the version control system they were using; apparently 
the system was designed (and purchased) for a much smaller number of 
projects and consequently a much smaller number of changes than what was 
by now happening at that company, and was now running out of IDs for the 
changes. The software company provided support by shipping a 
custom-tailored build of the software using larger data fields for those 
IDs.

Also, there are types of software - most notably any non-trivial 
business software - where unanticipated operating conditions are the 
norm; no two companies of any significant size have the same business 
processes. Therefore, standard business software is designed to be 
adapted for each individual customer, but this adaptation process is 
non-trivial because of the flexibility the software offers.

Some software companies seem to be able to provide this level of support 
free of charge, but I have no problem with software companies covering 
the costs for this type of post-sale development by selling service 
agreements.

But of course if you define support := user-driven on-demand training of 
individual end users in how the software is supposed to work, then yes, 
if that's your business model then you're certainly doing something wrong.

Post a reply to this message

On 8/1/2015 12:09 AM, Jim Henderson wrote:
> On Fri, 31 Jul 2015 19:47:14 +0100, Orchid Win7 v1 wrote:
>
>> Selling "support" also works. Provided you have the staff to actually
>> deliver support.
>
> The problem with this is that if you're selling support, your product is
> poorly designed and difficult to use.
>
> If you're a software company that depends on support and training as
> bottom-line revenue figures, you're doing software wrong IMHO.
>

To add to Francois and Clipa’s comments (I hope I got the apostrophe right).
ERP software is initially configured with the help of the customer’s 
management and key users. They know what they want and outline their 
requirements. By the time the blueprint is designed, and implemented. 
The customer knowns more about how the software works and often redefine 
their requirements. This can extend the implementation to the extent it 
will never be finished. (At BASF the German branch took 10 years on a 
pilot project and they were still not finished.) So there is generally a 
“Freeze” where there are no more changes allowed. After the “Go Live” 
the support company will make the changes wanted. This can be quite 
extensive.
These programs are quite complex and it is not normal for the company to 
do them as they generally do not have the expertise. If the company is 
regulated by, say the FDA, Financial authorities, Nuclear regulators 
etc. It has to be done by an authorised external company.
User support is almost a freebie.



-- 

Regards
     Stephen

Post a reply to this message

On Sat, 01 Aug 2015 10:37:20 -0400, Francois Labreque wrote:

> Le 2015-07-31 19:09, Jim Henderson a écrit :
>> On Fri, 31 Jul 2015 19:47:14 +0100, Orchid Win7 v1 wrote:
>>
>>> Selling "support" also works. Provided you have the staff to actually
>>> deliver support.
>>
>> The problem with this is that if you're selling support, your product
>> is poorly designed and difficult to use.
>>
>> If you're a software company that depends on support and training as
>> bottom-line revenue figures, you're doing software wrong IMHO.
>>
>>
> You're saying a oil-rig management system, or an MRI control system
> shouldn't require training?

Of course not.  There are things relating to the system that are not 
about the system itself.

What I am saying is that if your UI is so complex that it needs a guide 
or a training class, then your UI is not properly designed.

> I agree that if you're selling 3D-tictactoe-as-a-Service, and expect
> people to call you for support, you may have a problem, but complex
> software such as ERPs, or even industry-grade CAD systems _should_
> require some level of training to install and operate properly.

No, training is required for the job.  In order to use a CAD system to 
design aircraft properly, for example, you need to understand aircraft 
design.

Using the UI should be intuitive to someone with the necessary skills.

Jim

-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On Sat, 01 Aug 2015 18:21:26 +0200, clipka wrote:

> Support := help with operating the software in conditions not
> anticipated during the design.

In software these days, design tends to follow implementation - which is 
backwards.

That's what leads to a lot of broken UIs.  No design before 
implementation - the design comes with the implementation, and it follows 
the implementation rather than having a UX plan before the implementation 
starts.

> Nothing wrong with that in my book, provided the company isn't shifting
> costs from the design phase to the support phase.

Indeed, that's the problem I'm talking about - but with the added bit of 
"pay us to tell you how to use it properly".

Jim



-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On Sat, 01 Aug 2015 19:29:40 +0100, Stephen wrote:

> By the time the blueprint is designed, and implemented. The customer
> knowns more about how the software works and often redefine their
> requirements. This can extend the implementation to the extent it will
> never be finished.

Very true.

Two words you mention at the start, though, are not often enough included 
in software development:  "blueprint" and "design".

Specifically, user interaction design.

Jim



-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

Am 03.08.2015 um 02:50 schrieb Jim Henderson:
> On Sat, 01 Aug 2015 18:21:26 +0200, clipka wrote:
>
>> Support := help with operating the software in conditions not
>> anticipated during the design.
>
> In software these days, design tends to follow implementation - which is
> backwards.

No, actually that's where the software industry started. The truth is 
that things have gotten a tad better since then.

(Except in the domain of Open Source Software, where development is 
still driven by the developers rather than the users.)

> That's what leads to a lot of broken UIs.  No design before
> implementation - the design comes with the implementation, and it follows
> the implementation rather than having a UX plan before the implementation
> starts.

This thread started about Microsoft, didn't it?

You're certainly looking in the wrong direction there. Just look at 
Office 2010, and the loads of UI analysis and research went into it. Or 
Microsoft's primary programming language and environment, Visual Studio 
and C#, which in my book is as close as anyone has ever gotten to a 
programmer's dream.

Post a reply to this message

On Mon, 03 Aug 2015 09:10:00 +0200, clipka wrote:

> Am 03.08.2015 um 02:50 schrieb Jim Henderson:
>> On Sat, 01 Aug 2015 18:21:26 +0200, clipka wrote:
>>
>>> Support := help with operating the software in conditions not
>>> anticipated during the design.
>>
>> In software these days, design tends to follow implementation - which
>> is backwards.
> 
> No, actually that's where the software industry started. The truth is
> that things have gotten a tad better since then.

Yes, that is where it started, but it's still a poor way to build 
products.

> (Except in the domain of Open Source Software, where development is
> still driven by the developers rather than the users.)

Very much so.

>> That's what leads to a lot of broken UIs.  No design before
>> implementation - the design comes with the implementation, and it
>> follows the implementation rather than having a UX plan before the
>> implementation starts.
> 
> This thread started about Microsoft, didn't it?
> 
> You're certainly looking in the wrong direction there. Just look at
> Office 2010, and the loads of UI analysis and research went into it. Or
> Microsoft's primary programming language and environment, Visual Studio
> and C#, which in my book is as close as anyone has ever gotten to a
> programmer's dream.

I'm talking about interaction design, not design done by developers.

An interesting read - The Inmates are Running the Asylum, by Alan Cooper 
- describes the problem more fully.

Jim
-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA256

Le 03/08/2015 09:10, clipka a écrit :
> Office 2010, and the loads of UI analysis and research went into
> it.

When you are used to WinWord since 1995, Office 2010 is a PITA for
usual edition: the style short-cut/rolling selector does not list all
the available style, you have to expand it full size (as a size panel)
and it's not the same list. And the menus are all
"redesigned/reorder"... Reorganising windows of 2 documents is
difficult to have them side by side (and it engages "sync-scroll",
always, which is  good for different versions but suck for document
that are not identical but you want to read side by side (such as High
level & low level design))... 3 windows side by side ? forget it, no
more possible (or give me the way to do it: yes, stacked you can have,
but not left/center/right, using the mouse would be cheating )

And installation sticks to activate auto-correction and other
misbehaviour. (I know how to get «, I do not want Word to replace my
input of " with « because it's more fancy... especially when it's
code). And auto-capitalisation of first letter at start of sentence is
bad also for code.

The docx format is painful also: being a zipped xml, but without
internal management of versions, it is hardly compatible with any DCMS
(well, unless each version/commit become a blob by itself, with the
associated problems of disk space and bandwidth).


-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iJwEAQEIAAYFAlW/js0ACgkQhKAm8mTpkW0ICAP/ew0UtZKcuxUBzUwZrilmbuhA
fgpbMWQWbJKAcU+pMpzvj2vRvQOpu5dMdPX3zyOnmtiaN/fMsU08uY749qW0N3dz
7eMPqdr4n0ePbBo+OYVLMawX9MAMsDhjaOqxwKkfpzayrFO+5/GubhtVtEv6ubBa
hSYlT4zw2i95hA3cl04=
=emCB
-----END PGP SIGNATURE-----

Post a reply to this message

Am 03.08.2015 um 17:51 schrieb Jim Henderson:

>>> That's what leads to a lot of broken UIs.  No design before
>>> implementation - the design comes with the implementation, and it
>>> follows the implementation rather than having a UX plan before the
>>> implementation starts.
>>
>> This thread started about Microsoft, didn't it?
>>
>> You're certainly looking in the wrong direction there. Just look at
>> Office 2010, and the loads of UI analysis and research went into it. Or
>> Microsoft's primary programming language and environment, Visual Studio
>> and C#, which in my book is as close as anyone has ever gotten to a
>> programmer's dream.
>
> I'm talking about interaction design, not design done by developers.

And that's what I'm talking about as well, especially with regards to 
Office 2010. (With VS and C# it's a bit blurry, as its users /are/ 
developers after all; still, the users are not the developers of the 
piece of software we're discussing. So when I call it a programmer's 
dream, in this sense it's a user's dream, not a developer's dream.)

Unless I'm utterly misunderstanding what you mean by "interaction design".

Post a reply to this message

On 8/3/2015 6:46 PM, clipka wrote:
> Unless I'm utterly misunderstanding what you mean by "interaction design".

He is in double huff on a diagonal. So it is not surprising. ;-)

-- 

Regards
     Stephen

Post a reply to this message

Am 03.08.2015 um 17:54 schrieb Le_Forgeron:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> Le 03/08/2015 09:10, clipka a écrit :
>> Office 2010, and the loads of UI analysis and research went into
>> it.
>
> When you are used to WinWord since 1995, Office 2010 is a PITA for
> usual edition:

... and that tradition of poorly (if at all) designed UI is exactly what 
the developers of Office 2010 wanted to do away with.

I don't like working with Office 2010 myself, for exactly the same 
reason; still, you can't argue with two fact: (1) Microsoft put a hell 
lot of effort into analyzing user expectation and interaction during the 
design of the Office 2010 interface; and (2) for people /not/ accustomed 
to the - let's be honest - horrible user interface of prior versions of 
MS Office, the Office 2010 user interface is probably quite a good one.

> The docx format is painful also: being a zipped xml, but without
> internal management of versions, it is hardly compatible with any DCMS
> (well, unless each version/commit become a blob by itself, with the
> associated problems of disk space and bandwidth).

As if the prior binary file formats were any better for version 
management...

Also, it would be much easier in terms of version management if DCMS had 
learned to deal with zipped files already. It's a pretty standard way of 
keeping files (or sets thereof) small, so why the Bloops haven't DCMS 
learned to unpack them long ago?

In my book, that's poor design of DCMS rather than the Office file 
format. As a matter of fact, if it wasn't for the zipping I guess the 
DOCX file format would make for a pretty easy format to version manage.

Post a reply to this message

On Mon, 03 Aug 2015 19:46:08 +0200, clipka wrote:

> Am 03.08.2015 um 17:51 schrieb Jim Henderson:
> 
>>>> That's what leads to a lot of broken UIs.  No design before
>>>> implementation - the design comes with the implementation, and it
>>>> follows the implementation rather than having a UX plan before the
>>>> implementation starts.
>>>
>>> This thread started about Microsoft, didn't it?
>>>
>>> You're certainly looking in the wrong direction there. Just look at
>>> Office 2010, and the loads of UI analysis and research went into it.
>>> Or Microsoft's primary programming language and environment, Visual
>>> Studio and C#, which in my book is as close as anyone has ever gotten
>>> to a programmer's dream.
>>
>> I'm talking about interaction design, not design done by developers.
> 
> And that's what I'm talking about as well, especially with regards to
> Office 2010. (With VS and C# it's a bit blurry, as its users /are/
> developers after all; still, the users are not the developers of the
> piece of software we're discussing. So when I call it a programmer's
> dream, in this sense it's a user's dream, not a developer's dream.)
> 
> Unless I'm utterly misunderstanding what you mean by "interaction
> design".

Interaction design = design that implements features that facilitate 
useful user interaction, rather than features that are focused on "we 
implemented this feature, and here's an interface to use it".

For example, if you have an application that protects web resources, the 
interface needs to facilitate protecting web resources - it should not 
focus on configuring individual objects that are used to protect those 
resources, and leave it to the user to figure out how they are related to 
each other.

Tie idea is that there needs to be some elegance and simplicity in the 
design.  *Most* software "design" is done during development, rather than 
preceding it, and so the form follows the interface rather than designing 
how the interface workflow should work, and then using that as 
scaffolding for the underlying code that takes care of the details.

Jim

-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On 03/08/2015 08:10 AM, clipka wrote:
> Or
> Microsoft's primary programming language and environment, Visual Studio
> and C#, which in my book is as close as anyone has ever gotten to a
> programmer's dream.

Um... hello there. :-}

Post a reply to this message

On 8/3/2015 7:46 PM, Jim Henderson wrote:
> Interaction design = design that implements features that facilitate
> useful user interaction, rather than features that are focused on "we
> implemented this feature, and here's an interface to use it".
>
> For example, if you have an application that protects web resources, the
> interface needs to facilitate protecting web resources - it should not
> focus on configuring individual objects that are used to protect those
> resources, and leave it to the user to figure out how they are related to
> each other.
>
> Tie idea is that there needs to be some elegance and simplicity in the
> design.*Most*  software "design" is done during development, rather than
> preceding it, and so the form follows the interface rather than designing
> how the interface workflow should work, and then using that as
> scaffolding for the underlying code that takes care of the details.


I think I disagree with that concept.
For me, education is King or Queen. (I am an equal opportunity know it all.)
When you start simplifying complex software to the extent you think the 
man on the Clapham omnibus can operate it without any training. You are 
doing no one any favours. I learnt a word recently. It is nerfed. And 
that is the Micro$oft way.
Form should follow function, not the other way around.
IMO

-- 

Regards
     Stephen

Post a reply to this message

Am 03.08.2015 um 21:47 schrieb Stephen:

> I think I disagree with that concept.
> For me, education is King or Queen. (I am an equal opportunity know it
> all.)
> When you start simplifying complex software to the extent you think the
> man on the Clapham omnibus can operate it without any training. You are
> doing no one any favours. I learnt a word recently. It is nerfed. And
> that is the Micro$oft way.
> Form should follow function, not the other way around.
> IMO

Good old Bauhaus tradition.

On the other hand, the Bauhaus designers meant something different when 
referring to "form" and "function". To them, "function" included 
ergonomic requirements (such as, you should be able to hold an electric 
razor in one hand conveniently, and the power cord should be placed in 
such a way that you won't strangle yourself with it), while "form" was 
strictly limited to aesthetic aspects.

I think one branch of software development that's doing a pretty good 
job in terms of User Experience is the gaming industry. Look at what 
they do when it comes to graphics configuration: They usually provide 
you with a simple 1-dimensional "quality" setting to balance the 
graphics quality vs. speed, hiding the technical details of what each 
setting actually means in technical terms. At the same time they do 
provide an additional interface for people who think they know better, 
allowing them to tweak all the little details. (And often there's even a 
third tier of technical tweakables, for which there is no user interface 
except a config file.)

Post a reply to this message

Am 03.08.2015 um 21:35 schrieb Orchid Win7 v1:
> On 03/08/2015 08:10 AM, clipka wrote:
>> Or
>> Microsoft's primary programming language and environment, Visual Studio
>> and C#, which in my book is as close as anyone has ever gotten to a
>> programmer's dream.
>
> Um... hello there. :-}

Don't say a word. You're a Haskell enthusiast, and therefore by 
definition you don't have programmer's dreams - you have psychotic 
hallucinations :-P

Post a reply to this message

On Mon, 03 Aug 2015 20:47:49 +0100, Stephen wrote:

> On 8/3/2015 7:46 PM, Jim Henderson wrote:
>> Interaction design = design that implements features that facilitate
>> useful user interaction, rather than features that are focused on "we
>> implemented this feature, and here's an interface to use it".
>>
>> For example, if you have an application that protects web resources,
>> the interface needs to facilitate protecting web resources - it should
>> not focus on configuring individual objects that are used to protect
>> those resources, and leave it to the user to figure out how they are
>> related to each other.
>>
>> Tie idea is that there needs to be some elegance and simplicity in the
>> design.*Most*  software "design" is done during development, rather
>> than preceding it, and so the form follows the interface rather than
>> designing how the interface workflow should work, and then using that
>> as scaffolding for the underlying code that takes care of the details.
> 
> 
> I think I disagree with that concept.
> For me, education is King or Queen. (I am an equal opportunity know it
> all.)
> When you start simplifying complex software to the extent you think the
> man on the Clapham omnibus can operate it without any training. You are
> doing no one any favours. I learnt a word recently. It is nerfed. And
> that is the Micro$oft way.
> Form should follow function, not the other way around.
> IMO

Take a look at Apple products and interfaces, then take a look at 
Microsoft products and interfaces.

Apple understands the benefits of designing before you implement the 
backend.

The trick is to not dumb down the capabilities, but to make them easy to 
use.

Jim



-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On Mon, 03 Aug 2015 23:44:27 +0200, clipka wrote:

> Am 03.08.2015 um 21:47 schrieb Stephen:
> 
>> I think I disagree with that concept.
>> For me, education is King or Queen. (I am an equal opportunity know it
>> all.)
>> When you start simplifying complex software to the extent you think the
>> man on the Clapham omnibus can operate it without any training. You are
>> doing no one any favours. I learnt a word recently. It is nerfed. And
>> that is the Micro$oft way.
>> Form should follow function, not the other way around.
>> IMO
> 
> Good old Bauhaus tradition.
> 
> On the other hand, the Bauhaus designers meant something different when
> referring to "form" and "function". To them, "function" included
> ergonomic requirements (such as, you should be able to hold an electric
> razor in one hand conveniently, and the power cord should be placed in
> such a way that you won't strangle yourself with it), while "form" was
> strictly limited to aesthetic aspects.
> 
> I think one branch of software development that's doing a pretty good
> job in terms of User Experience is the gaming industry. Look at what
> they do when it comes to graphics configuration: They usually provide
> you with a simple 1-dimensional "quality" setting to balance the
> graphics quality vs. speed, hiding the technical details of what each
> setting actually means in technical terms. At the same time they do
> provide an additional interface for people who think they know better,
> allowing them to tweak all the little details. (And often there's even a
> third tier of technical tweakables, for which there is no user interface
> except a config file.)

Exactly. :)

Jim



-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message

On 8/3/2015 10:48 PM, clipka wrote:
> Am 03.08.2015 um 21:35 schrieb Orchid Win7 v1:
>> On 03/08/2015 08:10 AM, clipka wrote:
>>> Or
>>> Microsoft's primary programming language and environment, Visual Studio
>>> and C#, which in my book is as close as anyone has ever gotten to a
>>> programmer's dream.
>>
>> Um... hello there. :-}
>
> Don't say a word. You're a Haskell enthusiast, and therefore by
> definition you don't have programmer's dreams - you have psychotic
> hallucinations :-P
>

Don't beat about the bush, Clipka. Tell him what you really think. ;-)

-- 

Regards
     Stephen

Post a reply to this message

On 8/4/2015 12:36 AM, Jim Henderson wrote:
>> I think I disagree with that concept.
>> >For me, education is King or Queen. (I am an equal opportunity know it
>> >all.)
>> >When you start simplifying complex software to the extent you think the
>> >man on the Clapham omnibus can operate it without any training. You are
>> >doing no one any favours. I learnt a word recently. It is nerfed. And
>> >that is the Micro$oft way.
>> >Form should follow function, not the other way around.
>> >IMO

> Take a look at Apple products and interfaces, then take a look at
> Microsoft products and interfaces.
>
> Apple understands the benefits of designing before you implement the
> backend.
>

I spent a whole five minuets thinking about this. And have come to the 
conclusion that I have turned into the older generation, already.

Physically Apple products beat everything else hands down. IMO
Interfaces, they are not intuitive to me, too much thought has gone into 
them and I feel that they are over engineered. But then when technology 
morphs into consumer products. Something has to change to let the little 
darlings use it without straining their capabilities.
So I will sit back, keep my gob shut, if I can, and wait for the sky to 
fall.


> The trick is to not dumb down the capabilities, but to make them easy to
> use.

I could not agree more. It is the detail where the devil resides.
I would like to hear Patrick's opinion on this.
-- 

Regards
     Stephen

Post a reply to this message

On 04/08/2015 06:53 AM, Stephen wrote:
> On 8/3/2015 10:48 PM, clipka wrote:
>> Don't say a word. You're a Haskell enthusiast, and therefore by
>> definition you don't have programmer's dreams - you have psychotic
>> hallucinations :-P
>>
>
> Don't beat about the bush, Clipka. Tell him what you really think. ;-)

Don't worry, I already got that from this guy:

http://www.theregister.co.uk/2012/12/21/financial_software_disasters/?page=2

"And one language I’ve been warned about, though I’ve never had the 
opportunity to use it, is Haskell, an offshoot of ML. According to a 
friend in academia who’s studied it, it’s “the Taliban version of ML,” 
in which it’s all but impossible to write readable code."

Yes, because if you read the linked Stack Overflow question, you'll see 
how this function is actually "x = 1 : map (*2)", followed by no less 
than 15 steps of deliberate obfuscation.

In other words "I looked up the most unnecessarily complex piece of code 
I could find in language XYZ, thereby proving that it's impossible to 
write readable code in language XYZ".

Jesus, just because a language requires you to *use your brain* and 
learn to *think differently* does not mean it is "impossible to write 
readable code with it". >:-[

Post a reply to this message

> To say nothing of the privacy and confidentiality issues of having
> Microsoft have access to every file you ever create. (I doubt too many
> corporate types would like having their propriety data on a hostile 3rd
> party server.)

That's why I say medium-large corporations will be the problem. They 
have set, highly evolved ways of doing things (including security) and 
will be reluctant to change.

>> Their big problem will be the medium-large corporations that take
>> months, if not years to test and roll out major software updates. There
>> is no way they would accept the possibility of one day their entire
>> company coming to a halt with millions of pounds lost due to an MS
>> "update" that has broken something somewhere within their business. Also
>> a lot of systems are not connected to the internet for various reasons,
>> how would they work?
>
> They also have a problem with SOHO setups where people wouldn't know
> what "computer security" is if it hit them in the face.

I wouldn't be surprised if having all your files on a remote MS server 
somewhere woudl actually be *more secure* than left on someone's SOHO 
setup that has no clue about security.

Post a reply to this message

> I agree that if you're selling 3D-tictactoe-as-a-Service, and expect
> people to call you for support, you may have a problem, but complex
> software such as ERPs, or even industry-grade CAD systems _should_
> require some level of training to install and operate properly.

I get the distinct impression that some software (especially >£10k) is 
kept more complex than it needs to be *on purpose* so that companies can 
sell expensive support contracts.

Post a reply to this message

On 8/4/2015 8:19 AM, Orchid Win7 v1 wrote:
>> Don't beat about the bush, Clipka. Tell him what you really think. ;-)
>
> Don't worry, I already got that from this guy:
>
> http://www.theregister.co.uk/2012/12/21/financial_software_disasters/?page=2
>
>

Thanks for the link. I enjoyed reading it because it confirms my 
prejudices. :)

> “They think that if they’re not writing 80 lines of code to add two numbers,
they’re not using their education.”

How true that one is.
In SAP there is a trend for programmers to move into configuration. 
Their implementations are all about writing bespoke code to do what can 
be done with half an hour's training or a small change in methodology.
Nightmare!

> "And one language I’ve been warned about, though I’ve never had the
> opportunity to use it, is Haskell, an offshoot of ML. According to a
> friend in academia who’s studied it, it’s “the Taliban version of ML,”
> in which it’s all but impossible to write readable code."
>
> Yes, because if you read the linked Stack Overflow question, you'll see
> how this function is actually "x = 1 : map (*2)", followed by no less
> than 15 steps of deliberate obfuscation.
>
> In other words "I looked up the most unnecessarily complex piece of code
> I could find in language XYZ, thereby proving that it's impossible to
> write readable code in language XYZ".
>

Well, what do you expect when someone wants to make a point? ;-)

> Jesus, just because a language requires you to *use your brain* and
> learn to *think differently* does not mean it is "impossible to write
> readable code with it". >:-[

Do you think that this is because that using your brain is a threat to 
your management?

Remember (no it is too long ago to actually remember) education for the 
working class and slaves. Was considered a bad thing for the ruling 
classes. It gave the lower orders ideas above their station.

-- 

Regards
     Stephen

Post a reply to this message

On 8/4/2015 8:50 AM, scott wrote:
>> I agree that if you're selling 3D-tictactoe-as-a-Service, and expect
>> people to call you for support, you may have a problem, but complex
>> software such as ERPs, or even industry-grade CAD systems _should_
>> require some level of training to install and operate properly.
>
> I get the distinct impression that some software (especially >£10k) is
> kept more complex than it needs to be *on purpose* so that companies can
> sell expensive support contracts.
>


Wheesht! ;-)

-- 

Regards
     Stephen

Post a reply to this message

On 8/3/2015 10:44 PM, clipka wrote:

>
> Good old Bauhaus tradition.
>

Indeed! So good that the Nazis shut them down.
That is a plus in my opinion. :-)

> On the other hand, the Bauhaus designers meant something different when
> referring to "form" and "function". To them, "function" included
> ergonomic requirements (such as, you should be able to hold an electric
> razor in one hand conveniently, and the power cord should be placed in
> such a way that you won't strangle yourself with it), while "form" was
> strictly limited to aesthetic aspects.
>

True. But different times and different industries.

> I think one branch of software development that's doing a pretty good
> job in terms of User Experience is the gaming industry. Look at what
> they do when it comes to graphics configuration: They usually provide
> you with a simple 1-dimensional "quality" setting to balance the
> graphics quality vs. speed, hiding the technical details of what each
> setting actually means in technical terms. At the same time they do
> provide an additional interface for people who think they know better,
> allowing them to tweak all the little details. (And often there's even a
> third tier of technical tweakables, for which there is no user interface
> except a config file.)
>

I don't have much experience outside of Elite.
What you wrote describes how it is. A good thing too. :-)
Where would you put addons like SweetFx? Which BTW I cannot get to work.

-- 

Regards
     Stephen

Post a reply to this message

> Then too, at what point do you think that the governments of the world will
> become comfortable with their internal communications being stored -- even
> temporarily -- in the buffer of someone else's computers? Personally, I think
> they'll go back to using typewriters first, but realistically, they'll probably
> just switch to linux.

More realistically, MS would never take any decision that forced a 
significant proportion of its user base to switch away from Windows.

Post a reply to this message

On Tue, 04 Aug 2015 08:50:02 +0100, scott wrote:

>> I agree that if you're selling 3D-tictactoe-as-a-Service, and expect
>> people to call you for support, you may have a problem, but complex
>> software such as ERPs, or even industry-grade CAD systems _should_
>> require some level of training to install and operate properly.
> 
> I get the distinct impression that some software (especially >£10k) is
> kept more complex than it needs to be *on purpose* so that companies can
> sell expensive support contracts.

There is a cynical part of me that thinks that as well.

Jim
-- 
"I learned long ago, never to wrestle with a pig. You get dirty, and 
besides, the pig likes it." - George Bernard Shaw

Post a reply to this message