Reaping the Benefits of Cloud Computing

Posted by challisc PM | on Wed, 17/11/2010 - 11:28  1399  23 comments | report

In times gone by everything was on paper. It wasn’t practical to copy incoming  correspondence. Documents issued were often in triplicate, with at least two copies kept internally in different files and locations.

Along came on-premise computing, initially to automate processes like sales invoicing, then management information, plus other benefits such as electronic document management and workflow for incoming correspondence. Improvements in telecoms meant multiple sites could be networked, and remote access given to staff and selected business partners. In each case only if you could afford it.

Yes there’s been dozens of hi-profile cases of failure of on-premise systems, in selection, implementation or administration. There’s also plenty of cases that don’t get reported. For example I was discussing a computer audit report with the IT director at a well known electronics company some years ago when all three warning lights lit up on the mainframe room’s wall.  He barely flinched, but someone had cut through the datacomms and network cables when replacing the voice system. All contact had been lost with the warehouse, the worldwide network, and even terminals in the same building. Business stopped. The easiest sale of a disaster recovery planning exercise ever!

Best Practice in On-Premise Computing

Over the years “best practice” with on-premise computing has developed so risks can be managed sensibly and cost-effectively. Indeed I can’t think of a case of on-premise failure that isn’t due to not following “best practice” in selection, implementation or administration. That includes failure of part of the procedures, of which failure of an untested backup is probably the most common. There is always a residual risk. But as for any insurance, insurers will typically insist on applying best endeavours, i.e. best practice, to keep risks to a minimum.

The Benefits of Cloud Computing

Along comes cloud computing, which for most of us is applications in the form of SaaS (Software as a Service, formerly known as Application Software Provision, ASP). This can provide several significant benefits, including:

  1. Contingency, back-up and recovery facilities that are far superior to what is typically achieved in many organisations. (However I’ve heard of at least one IT outsource specialist who has suffered from backup failure.)
  2. Remote access and other functionality that is far cheaper than could be achieved in-house. In many cases that opens up new possibilities that are not affordable on an in-house basis, for smaller and larger organisations alike.
  3. Easier to pilot-test and quicker to acquire than on-premise systems (though training, data conversion and other aspects of implementation will be similar for a system of similar complexity and ease of use)
  4. Reduction of in-house administration, reducing hassle and cost

Handling Pitfalls and Risks of Cloud Computing

However cloud computing introduces new pitfalls and risks, which if not adequately covered stops us benefiting from the cloud either now or if something goes wrong later. “Best practice” is still in its infancy and evolving, and the imminent launch of the Code of Practice from the Cloud Industry Forum (CIF) is a great step. However by necessity the Code suggests what aspects need to be covered by providers, not necessarily how to do so, and will continue to evolve.

There are also aspects which are in the end-users court, such as duplicating internet access with a second method.

One issue that crosses my mind is going concern in audit reports. Detailing key risks has long been a disclosure requirement when listing on a stock exchange, and very often appears in business plans where money is otherwise being raised.  Dependency on one customer or supplier is an issue, as of course is dependency in key areas on computing, be it on-premise or cloud.

The production of CIF’s Code highlights what I have felt about the cloud industry for some years:

  • That only the better players are suitable for business critical applications. This is especially true if the end-user organisation is so dependent on the system that the organisation’s very existence may be at stake. In the on-premise world, loss of the receivables information has been shown likely to trigger collapse. As people become dependent on cloud systems for more and more of their operations, that risk widens.
  • Usage would increase if end-users could be more confident about the issues, to the obvious benefit of the industry

So How Can This Forum Help?

This forum is called a “discussion group”. We shouldn’t forget the benefits of the cloud, which of course is why we’re interested. But I see this forum as a place to discuss, and discussion is inevitably around issues that concern people. I hope the end result is an increase in the use of cloud computing, when we can adequately address the issues.

People will always take risks, whether it is not wearing a car seat belt, not insuring a house, or cutting corners with computing. That’s up to you, but better to know that there is a corner.

I was an early adopter of cloud computing, over 10 years ago. Fingers have been burnt. My own involvement in this forum is much more about self-training, with marketing only a bonus. I have been asking questions where I’m looking for better answers, and several tips and useful insights have emerged. I want to more safely use cloud computing in my own businesses, and more happily recommend cloud computing to clients, subject to the providers for a specific app.

How individual providers address the issues raised in this Forum is therefore an opportunity for them to show their competitive advantage, as some have usefully done in earlier discussions. Or otherwise it’s an opportunity for providers to better understand their customers’ concerns, and think how they can improve their offering or the way it is portrayed. 

Over To You

So without needing to focus on any one issue, now’s a chance to discuss how we as end-users and providers want to use this Forum. An opportunity to let off steam, if you wish.

A side-issue is that when I took over the manager position, we wanted to develop this Forum into industry and other apps rather than just those relevant to accounting practices. This is because a lot of the issues are generic across all apps and sizes of business. But recent responses suggest there are specific issues for practitioners that possible justify a separate Forum. I’m happy to start a new one with a wider remit, with perhaps a practitioner running the existing one.

Your views welcomed, with of course the usual professional courtesy should there be any disagreement. Debate is good!

Comments
chanpangchi's picture

My 2 cents -

chanpangchi | Wed, 17/11/2010 - 14:17 | Permalink

I want to add some more benefits to your great artcile:

  • Optimized Cashflow - since we don't need to invest on hardware and software and we don't need to worry about future software upgrade and potential hardware failure, our IT cost is more predictable and manageable.
  • Faster Return - Clound implementation typically take shorter implementation period, so faster return.
  • Better Scalability - we can double, triple our usage within a short period of request.

 

-- Regards,

chan_a@algconsultings.com

http://ca.linkedin.com/in/alginc

http://www.algconsultings.com/

challisc's picture

Faster Return

challisc | Thu, 18/11/2010 - 19:30 | Permalink

Thanks for the additional benefits Andrew (@chanpangchi). Cloud computing can certainly provide a faster return. But as we've discussed, this is through quicker acquisition than actual implementation. Virtually every implementation step such as data conversion, system integration and training would require a similar time frame as an equivalent on-premise implementation.

There's therefore no inherent implementation advantage for an app being in the cloud. If the cloud app would be genuinely quicker to implement because it is simpler or easier to use, then it needs to be compared to on-premise options just like you would a simpler on-premise offering.

An app that leverages the cloud's advantages well is Customer Relationship Management (CRM), which has become the real success story of cloud. In part this is due to the absence of proper CRM systems in some businesses, but in other businesses is widespread dissatisfaction with existing CRM systems. Sales directors want instant solutions. CRM in the cloud is easy and cheap to "give it a try".

Having used an early CRM system without remote access, I know how useless the data is likely to be. But if the existing system has good remote access, any dissatisfication is likely to be because it has been poorly implemented. Then the new cloud CRM system is unlikely to perform any better unless the end-users pay better attention to implementation.

david_terrar's picture

Faster implemtation plus it's a shame there aren't more comments

david_terrar | Wed, 01/12/2010 - 10:36 | Permalink

Hi Chris,

Sorry Chris but you don't know what you are talking about - you're wrong on you're analysis of faster implementation.  With cloud solutions you approach the implementation in a different way - some of the steps are similar, but many are different.  None of the set up and installation tasks are required, and then it's usually a smaller project to prove it works, maybe with only 1 or 2 users.  Once proved, the rest of the project becomes a roll out of a working solution, and the nature of SaaS products (which have been influenced by consumer web apps and so have been desgned for signifcantly better ease of use) mean that less training is required, and there are generally better tools for support, self help, online documentation and the like.  We have 30 years of experience in implementing all manner of ERP, financial and supply chain management solutions of different shapes and sizes.  We switched to implementing cloud based solutions in 2004.  In that 6 years we reckon that the various types of project have needed somewhere between 50% and 10% of the implementation time for an equivalent on premise application.

There are case studies to corroborate our experience.  In addition, yesterday you were at the Business Cloud Summit.  During the Cloud HR customer panel the guy from Colt explained that he had implemented his Cloud HR solution 10 times quicker than the on premise applications he was used to.

Finally just wanted to comment that it's such a shame that this positive thread has attracted only a few comments, when the negative threads spawn 40 or 50 plus. 

David Terrar

www.d2c.org.uk and www.twinfield.co.uk

dgrant's picture

Implementation differences

dgrant | Wed, 01/12/2010 - 11:01 | Permalink

David,

You mention in your recent post that there are some steps of the implementation process that are the same and some that are not between onsite and SaaS applications.  Would you be able to elaborate at all or perhaps point me at a place where I can see and understand the differences. 

I too was at the cloud summit yesterday but unfortunately missed 'The Story of a Cloud', which may have answered some of my queries on implementation so be gentle if I have dropped a ball here.

Thanks  

david_terrar's picture

Clarifiction - apart from the missing ones, they are the same st

david_terrar | Wed, 01/12/2010 - 13:46 | Permalink

@dgrant,

I'm glad you've asked because I realize I was sloppy with the language (in my haste), so it's great to get the chance to clarify.  What I really meant was that, apart from the missing steps (like provisioning of hardware, installing software and other infrstructure set-up), some of the steps/activities might take the same amount of time as for an on premise application, but most actually take much less time.  There are no absolutes, and every system is different, but I'm tying to talk in terms of the typical on premise project vs. the typical SaaS application. For example there is a data migration step.  With some Cloud services that might take a similar time to an on premise application, but in many cases it's quicker because the Cloud provider has better tools and is geared up for regular data import/export in a way that on premise software providers probably aren't.  On premise and SaaS will both have a pilot stage, but for SaaS the pilot is often with 1 or 2 or a small handfull.  With on premise there has been a larger investment in hardware and software to set up the pilot sysem, and so it will usually involve a dozen or more people as part of the project justification. More people means more complexity and questions and more time.  Our experience is that the smaller SaaS pilot proves the system faster and gets to a result much quicker, and then you are in to roll out.  Let's look at the software itself.  In theory an on premise application could be desgned to be just as easy to use as the web apps we are used to, and have brilliant online documentation and self help support facilities.  In reality most aren't because they've been around for a long time using older technology.  On average most SaaS applications have been designed more recently in a better and more innovative way.  All of these things add up.  Every step/activity in the chain is generally shorter, and some (like setting up the next set of harware/software/infrastucture to support the next phase of the roll out) take zero time.

I'll try and find some case study material to support what I'm saying, but one of the complictions we've got is that there haven't been that many documented, and the ones that are around are 5 or more years old.  Typically these kinds of reports aren't cheap to produce, whiich is why there aren't that many around - the small SaaS players just can't afford to do it.  By the way, the Intellect SaaS Group has just started putting together a document of case studies to try and address this issue.  Should take a couple of months but I'll shout about it here when it's done.  I hope that helps.

David Terrar

www.d2c.org.uk and www.twinfield.co.uk

chanpangchi's picture

Let's look at the question from another angle - why don't we wan

chanpangchi | Wed, 01/12/2010 - 14:15 | Permalink

A lot of people told me that they were not comfortable about privacy / security.  This is a valid concern!

This is why we have to be careful about choosing our cloud provider.  Google, IBM, Microsoft, and Oracle would definitely provide higher level of privacy / security than most business.  They spend millions of dollars on security hardware; they have a whole team of secuirty acrchitects to look after their system and data.  They also have R&D to constantly look for better way improve security.  Most business just don't have the same resources.

There is no one can guarantee absolute privacy / security; we just have a very good demonstration from WikiLeaks yesterday.  However, let's ask ourselves the following questions.

How many of us are working for company which has a full time security architect?  What was the last year budget on IT security?  Do you have a privacy / security policy in place?

One of the company that I worked for, outsourced their whole infrastructure to IBM.  I was invited to their office for visit and one thing they showed off was their security infrastructure.  I was very impressed.

Can a non IT company do better in term of provacy / security than these IT giants?

What is your next concern not going to cloud?

-- Regards,

chan_a@algconsultings.com

http://ca.linkedin.com/in/alginc

http://www.algconsultings.com/

challisc's picture

Narrowing the gap

challisc | Wed, 01/12/2010 - 16:31 | Permalink

Thanks David for the reality check. Your subsequent clarification and examples were more consistent with what I was trying to say.

The remaining gap was that I didn’t originally write all I was thinking when I said “for a system of similar complexity and ease of use”. I was hoping others would say that a cloud system, written more recently with latest techniques and a more familiar browser-based front end, is typically easier to use for both end-users and implementers. But as you imply, the worst cloud app in this respect may or may not be as good as the best on-premise app in a specific niche.

As you’ve clarified, I was trying to make the distinction between acquisition (up to point the software is installed and ready to configure) and subsequent implementation. With a cloud system a low-cost pilot can be “acquired” today and scaled up tomorrow, whereas an on-premise system will take weeks or months to go through the acquisition steps of approval, procurement and installation. As you say, that can be more than once as a larger project rolls out. That accounts for much of the time saving in any cloud project, and is a major benefit. The other benefit of a cloud pilot is not having to make a major commitment to do the pilot should the app prove unsuitable.

But I’m not sure the corporate HR case study yesterday you referred to was entirely cloud-specific. Yes they benefited from the acquisition advantages. But it wasn’t clear to me that the remaining time savings were anything inherently to do with being in the cloud, although again may be more typical of it. What I heard was making use of the standard facilities in well-functioned software, without the months of complex configuration and modifications that he’d experienced with previous corporate implementations. I’ve seen very similar gains by taking the “make do” approach in the on-premise world. I suspect that had this software been available on-premise, most if not all the time gains would have been achieved there too. In any case in the mid-market and small business sectors, the on-premise approach is usually to make use of what’s in the app as standard, so there isn’t that type of excess time to be saved.

So yes, cloud apps are typically easier to use. If they are, they can provide time savings in the post-acquisition implementation phases, just as better on-premise software and implementation techniques can. Where cloud apps really score inherently is in cutting acquisition time, both initially and for any subsequent expansion.

One of the other case studies yesterday was the corporate implementation of a new cloud-based email system. The project leader said, if I heard him correctly, that he spent 90% of his time on training, so the system couldn’t have been totally intuitive. The reward was less support calls on the Monday after a weekend roll-out than they usually got from the users on the old system. Two points arise. Firstly the cloud-positive point that with he had less to think about in terms of technology and could concentrate on the users. Secondly that putting the right effort into implementation steps is still needed, and is rewarded with a system that works.

If you’ve got any other examples where an app is faster to implement inherently because it is in the cloud, we’d be interested to hear.

As to the number of comments, it seems that a post has had to be contentious to inspire debate. Alternatively as manager of this forum I’d encourage people to ask questions, whatever they are. Someone else is probably also asking themselves that question. A multitude of constructive answers would then be welcomed.

david_terrar's picture

You still don't quite get it

david_terrar | Wed, 01/12/2010 - 18:18 | Permalink

Hi Chris,

The Colt story didn't come across to me in the same way it did to you.  You talk about the "make do" approach to traditional software.  It has always been true that if you live with what the standard software does, or adapt your processes to the standard processes in the pakage, then you'll implement quicker, and you won't have to live with the maintenance headache of modified software and getting behind on versons. However, that is still different to the on premise vs SaaS comparison we are dicussing here.  As I said earlier, in theory the on premise solution could be as good as a new SaaS product, but in practice the jump between a Xero/Twinfield/Kashflow to a Sage (or SuccessFactors/Workday to PeopleSoft)  is quite a gap.

On your last point, I live in hope about the discussions getting more positive, and that the general accountant (in business or practice) who wants to find out about cloud isn't afraid to to ask questios here. This should be a resource to help the average AccountingWEB reader get to know Cloud accounting rather than just expert cloud advocates jousting wth expert cloud detractors (although we need both of those to fuel some debate).

David Terrar

www.d2c.org.uk an www.twinfield.co.uk  

 

It's the pregnant pauses, Andrew

Coopers Willy | Wed, 01/12/2010 - 18:25 | Permalink

Our firm (10 person practice) actively encourages clients into the cloud for their bookkeeping / CRM.  It's the only way to go!

Why, then, is our own firm's IT is still on-premises?  

I think that will gradually change, especially as all the really good new stuff seems mainly to be web based.  For us it'll probably be CRM and time recording and our own bookkeeping into the cloud first.  Not sure after that.

Ultimately I think it'll be great software that'll be the main catalyst for driving business into the cloud.  

Incidentally I had the IT guys who manage our network over today.  I asked him about the threat posed by the cloud to his business model.

He didn't seem worried.  Maybe he is relying on inertia.  But he said that he thinks the main problem with the cloud is speed and bandwidth, especially outside the big cities.  We can talk about security, etc, etc but at the end of the day it's that pregnant pause you experience after entering each transaction into a piece of SaaS software that puts people off.  And, he says, this is only going to get worse as improvements in the national IT infrastructure fails to keep up with increasing demand.  I think he's got a point.

david_terrar's picture

So good to hear that "it's the only way to go!"

david_terrar | Wed, 01/12/2010 - 19:08 | Permalink

Hi Coopers Willy,

You're right to remind us that multi-megabit broadband isn't universal, and that we have a digital divide in the UK, with some rural areas very poorly served.  However, you still shouldn't see much of a pregnant pause.  Some of the better cloud solutions that have been around for 5 or 10+ years have lived through a time when a 64k ISDN line was a really big deal and so have designed for the lower bandwidth of times gone by.  With a good Cloud solution (like good client/server software) you shouldn't notice the response time.

Very interested in your view on the switch and the priority sequence, but great to see you are encouraging your clients to adopt Cloud.   

David Terrar

www.d2c.org.uk and www.twinfield.co.uk

david_terrar's picture

Security from a different angle - you don't just have to look at

david_terrar | Sun, 05/12/2010 - 10:09 | Permalink

@chanpangchi

I've just re-read your contribution suggesting we look at this from a different angle, and I can't see how it helps someone trying to make sense of "reaping the Benefits of Cloud Computing".  Let me explain.

The key issue you are raising is to do with security, which is absolutely vital, but you are suggesting that "Google, IBM, Microsoft, and Oracle would definitely provide higher level of privacy / security than most business".  Yes the big boys can be good at this stuff, but so can the small providers.  If you talk to any of the significant online accounting players in the UK, many of them have less that 100, or 50, or 20 employees, but their systems are set up to be "hacker safe", they have failover and redundancy built in to their infrastructure, and they have a data backup strategy that most small businesses can't or don't afford - for example one I know encrypts and stores nightly data backups in 3 separate locations around the world - Europe, East Coast US and West Coast US - the kind of disaster recovery big corporations have in place.  Security is important but one of the huge benefits of the Cloud is that it's affordable for the little guy (whether buyer or seller).

Then when you talk about security you raise the topic of Wikileaks, which is completely irrelevant as a comparison to the Cloud security topic.  The latest leak of US Embassy Cables that you refer to were posted on a US government intranet, SIPDIS, for a very wide distribution among diplomatic, government and military circles.  There are around 3 million Americans cleared to read that material on that Intranet.  You might ask why they hadn't been leaked before?

Security is vital, but buyers shouldn't be scared in to thinking only large companies can provide it.

David Terrar

www.d2c.org.uk and www.twinfield.co.uk
 

daveforbes's picture

SIPRNET, physical security and errant employees

daveforbes | Sun, 05/12/2010 - 10:44 | Permalink

Why was it not leaked before ? To get to a connected PC you had to get past a burly chap with a gun.

david_terrar's picture

Physical security

david_terrar | Sun, 05/12/2010 - 11:15 | Permalink

daveforbes

Could have been  chapess, but I take your point.  :) 

Just changed this comment - thinkng about it, Wikileaks is a valid example of how difficult security can be for any size organization.  

David Terrar

www.d2c.org.uk and www.twinfield.co.uk   

Hacking HTTPS over Wi-Fi ...

JC | Sun, 05/12/2010 - 12:42 | Permalink

http://www.airtightnetworks.com/fileadmin/content_images/demos/yahoo_ssl_mkt/yahoo_ssl_mkt.html

 

daveforbes's picture

@JC

daveforbes | Sun, 05/12/2010 - 15:54 | Permalink

Watched your video. If you have just connected to a wifi that you know nothing about and you get a message box headed "Security Error : Domain name mismatch" and going on to say "You have attempted to establish a connection with "login.yahoo.com". However the security certificate presented belongs to "Anonymous User". It is possible though unlikely that someone may be trying to intercept your communication with this web site.", then Cancel is a more sensible option than OK.

The trouble is I suppose it only takes one moment of distraction and the damage is done.

chanpangchi's picture

Are you ready for winter?

chanpangchi | Sun, 05/12/2010 - 16:44 | Permalink

Hi David T,

Good point!

My point is to compare us (mainly SMB or business corpoarte) who may not have the same resources as those big IT guys to manage our security infrastructure.

But you are right, there are numerous cloud providers who may have the same ability to compete with the big guys.  When we select our cloud providers, we should investigate if they have something like:

  • International Organization for Standardization (ISO) 27001
  • Statement on Auditing Standards (SAS) 70 Type I and Type II
  • Health Insurance Portability and Accountability Act (HIPAA)
  • Family Educational Rights and Privacy Act (FERPA)
  • Title 21 CFR Part 11 of the Code of Federal Regulations
  • Federal Information Processing Standard (FIPS) 140-2
  • Trusted Internet Connections (TIC) compliance

We should also make sure if our professional body (Accountants, Actuaries, Financial Planners, Lawyers... etc) have any specific requirements on how to handle application / data / security.

Regarding Wikileaks, I just would like to use it to demonstrate how poor security can turn into a disasster.

I am a big supporter of cloud computing and I have virtually moved all my business applications to cloud.  I was very glad because my desktop's motherboard was dead about 2 weeks ago.  All I did was to plugin my keyboard, mouse and monitor to my laptop, then I could continue to work.  Have to admit my laptop is not as powerful as my desktop; so when I compiled my program, it would take longer time.  But apart this minor issue, I have no other problem. 

Are you ready for the winter?  Do you know how much productivity was lost due to severe weather each year?  If you to minimize the impact of severe weather, you may want to migrate to cloud.

BTW, you can call me Andrew.

-- Regards,

chan_a@algconsultings.com

http://ca.linkedin.com/in/alginc

http://www.algconsultings.com/

Total Cloud Computing

reghorman | Sun, 05/12/2010 - 20:09 | Permalink

Cloud benefits are significant and I'd like to draw the groups attention to being able to 'place' in the Cloud all current software used in a business. No need to change a thing. The Ultimate Cloud solution is the Virtual Desktop (and by definition Virtual Network). No need to change and Total Cloud offers the best value for money to maximise the impact and benefits of Cloud.

SaaS is Ok, but assumes ultility computing where change is 'forced' on users, that's OK for some uility software but will never reach the parts of a system that are uniqie to users. Placing what is already tried and tested in the Cloud to take advantage of the delivery benefits is available today. Users have been able to migrate in a few hours, transfer all software and data AND reduce other 'Component Cloud' service costs.

A virtual desktops delivers what you do today, anywhere and on any device and importantly is delivered through knowledge channels to ensure a human connection to the needs of business.

 

daveforbes's picture

@reghorman

daveforbes | Sun, 05/12/2010 - 21:06 | Permalink

I think for the purposes of these discussions "cloud" means applications that are web based rather than just accessing traditional software using a remote desktop via terminal services.  

chanpangchi's picture

What is cloud computing?

chanpangchi | Sun, 05/12/2010 - 21:44 | Permalink

Do we have a definition of what cloud computing is?

--

Regards,

Andrew Chan.

chan_a@algconsultings.com

http://ca.linkedin.com/in/alginc

http://www.algconsultings.com/

Definition ...

JC | Mon, 06/12/2010 - 09:14 | Permalink

@daveforbes - quite right web based

Why oh why do citrix, t/s based apps try to claim this monika

If you are citrix (et al) then just say so; it is a perfectly valid solution - just don't claim it's Cloud
 

challisc's picture

Working definition of Cloud Computing

challisc | Mon, 06/12/2010 - 09:56 | Permalink

As good a definition as any comes from The Cloud Industry Forum, being “any remotely hosted IT service of any type, including multi-tenanted systems accessed via the internet”.

Of most relevance to most businesses is SaaS (Software as a Service”). This covers business applications such as accounting or email which run on a third party’s servers, accessed via a browser, and paid for by some form of subscription for the service. 

But “cloud” doesn’t preclude other forms of access or payment schemes. There are also two other forms of cloud computing:

  • PaaS “Platform as a Service” which allows apps to be developed and deployed on a third party’s system
  • IaaS “Infrastructure as a Service” where you run your own apps on a third party’s hardware, who then also provide administration services such as backup and disaster recovery

There is also the concept of a “private cloud”, emulating cloud computing on private networks, but not necessarily involving a third party and remote hosting. As such this can fall outside the definition above, being only "cloud-like". Some organizations will use a mix of public and private cloud services, which is termed “hybrid cloud”.

Public SaaS using web browsers has and I’m sure will continue to be the focus of this forum, but is not the only form of cloud computing.
 

Mmmm ....

JC | Mon, 06/12/2010 - 11:14 | Permalink

Accept the CIF definition (don't necessarily agree with it)

But lets get more granular and explain how CIF makes the distinction between the different types of SaaS/Cloud - or is any differential regarded as irrelevant?

For instance how does a potential customer differentiate between the the latest crop of 'true' Cloud apps designed from the ground up to optimise this environment and the legacy offerings of many software house simply ported to this platform to gain marketing traction (i.e. running in an environment they were not originally designed for!)
 

david_terrar's picture

Standards, Cloud isn't just hosting, and definitions

david_terrar | Mon, 06/12/2010 - 11:34 | Permalink

Andrew,
Good to hear on all counts.  I'd already changed my mind on the Wikileaks thing - perfectly valid example of how difficult security is of any size organization.

On the principle of standards I agree with you, but on those particular standards I don't.  Things like the SAS 70 standard were all we had… but in general they are focused on technical issues that aren't that relevant to helping chose who is a good SaaS/Cloud supplier from who isn't.  Adopting them all would mean that only the large IT providers could play the game, and we need something that smaller Cloud providers can sign up to so buyers have a wide choice.  As Chris explains, the Cloud Industry Forum's Code of Practice is worth looking at - it's supported by the 3 UK SaaS/Cloud vendor groups and is governed by a cross section of vendors, customers, independents and lawyers.  Take a look at the other discussion thread on CIF for some more detail, but I believe that's a good answer for the UK.

@reghorman
I'm afraid I don't agree.  Taking what you currently have and hosting it on Cloud infrastructure will certainly do a job and give you some of the benefits of Cloud. The end of your argument about uniqueness would lead a company to only adopt bespoke systems rather than take advantage of the extra functionality packaged software can bring.  Most organizations, large or small, need a combination of both.  So your hosting only way would miss out on 3 things:
 ○ The extra benefits you can get from a true SaaS/Cloud approach such as the extra facilities that the provider can afford to add because of the 1 to  many economies of scale, the different approach to support of Cloud vs on premise, the continuous improvement that comes because the provider focuses on what customers are actually using the software for, etc. etc. (the real Cloud benefits)
 ○ The innovation that comes with moving to SaaS solution that has been designed recently and adopted the kind of approach we are used to in consumer apps (compared to the older on premise applications that are around)
 ○ The fact that you can chose a vendor like Salesforce or the others who have a Cloud based platform upon which you can develop the unique processes you need as well capitalizing on the true SaaS/Cloud benefits of the standard CRM, financial and other apps you need.

Accessing apps across the Internet using elastic Cloud Infrastructure is only part of the story.

Andrew, Chris, JC, all
If we have to get in to "what is Cloud" and Cloud definitions again (perfectly valid as the average business person still doesn't fully understand), can I suggest we start a new thread rather than have that discussion here.

David Terrar.

www.d2c.org.uk and www.twinfield.co.uk
 

Add comment
Log in or register to post comments
Group: Cloud Computing for Accountants discussion group
A place for accountants to share their thoughts about web-based systems
View all posts in this group
More posts from this group

Pages