University still living in the Nineties seeks help with move to SAP S/4HANA The UK's University of Leeds is looking for a systems integrator to help move its ERP system from an ageing SAP installation dating back to 1999 onto the German vendor's latest S/4HANA platform. The institution said it wanted to migrate away from its existing SAP ECC6 environment that has been in place from 1999. "Since then …
I was fortunate enough to leave Cambridge shortly before the CAPSA implementation, but I got regular updates from old colleagues as it all unfolded. The full report the Uni commissioned into it was published in the Cambridge University Reporter, and is online at:
https://www.admin.cam.ac.uk/reporter/2001-02/weekly/5861/
I do hope it's considered required reading by anyone considering somthing similar. It's a long read, but well worth the effort.
See also NHS, local authorities, police forces, shoe shops, car dealers...
Actually there is probably more common ground between similar faculties in different Universities than within one institution: science faculties vs arts faculties have very different resource/staffing/estates requirements for example.
That's why there are so many localisations and modifications layered on top of a general purpose business system.
I'm not sure "Shut up and eat your gruel" is quite the right marketing slogan though. If your product can't cope with all the wrinkles from the start, you're going to have to put in modifications. Or people will go back to bits of paper.
You could say the same about any sector.
Each institution is a business in it's own right. They have different ways of doing stuff and there have been plenty of arguments on El Reg about software vendors forcing organisation down sub-optimal routes.
On the other hand have spent many years working for such an institution I have also seen all the arguments that go round time and time again about the "requirements" that it appears no supplier can satisfy.
You cannot have these modifications that you had in the on-premises world, that's not how the cloud works.
Such a patronising statement. I thought the cloud is just someone else's computer that is supposed to have better uptime than the one in one's closet?
But this was inevitable - once big corporations take over the means of hosting, they'll dictate the terms of what you can't or can't run and what your organisation has to do.
It's a huge overstep.
Funny thing is that there is absolutely no reason for these apps to run on the "cloud" other than greed. Maybe regulators should look into that.
That's true some of SAP's own cloud offerings, but there's also other options to have S/4 in a cloudy environment and still have all the modifications. Of course SAP want you to move to one of their own cloudy as-a-service offerings, some of which you dance entirely to their tune, with minimal configuration options and constant pressure to subscribe to more more more.
It shouldn't be a surprise that any customisations created for ECC6 or other products will need some attention to work with or on S/4. And by attention I mean anywhere from review and test through to start again from scratch or switch to some equivalent standard feature instead.
Just like in the real world, there's a lot to be said for having a major cleanup whenever you move house. If nothing else it gives you more space to fill with cruft again.
"Funny thing is that there is absolutely no reason for these apps to run on the "cloud" other than greed. Maybe regulators should look into that."
Businesses have been led to believe that responsibility for all aspects of the application's operation devolve on the cloud provider ("whatever as a service"). Unfortunately they don't, as liability remains squarely with the client if something goes wrong but the client doesn't have control. It'll be very hard for regulators to step in except in specific cases, as it's essentially a matter of contract which is deemed to be an informed mutual agreement between the parties. Any contractual term that is not actually unlawful is hard to challenge onced entered into voluntarily.
They probably earn more than that now. Ive been out of the Blue Chip world (mostly) since 2006. Before I decided I hated working with "Blue Chip" companies, I was doing SAP integration contracting in the late 90s through to the mid 2000s I was getting £1200-1500 / day depending on the client (big pharma paid the best) and I would get mileage, hotels and expenses on top of that. I was only mid level and there were better skilled people earning upto £2500 / day. The rates were so disconnected from knowledge and expertise that it almost felt like theft at the time. It wasnt even particularly complicated compared to the embedded development world Im in now. The hardest parts of dealing with any SAP deployment are the politics first, then dealing with the fact that the sales team sold something that didnt or only partially existed
This post has been deleted by its author
"Since then, the environment has been added to, extending SAP and expanding the systems supporting HR, Finance and Purchasing with other third-party offerings"
The very definition, according to SAP, of an installation that is not going to work in S4.
Surely, if the existing system kinda-works, the correct move is to find a new system (not SAP) that does what the core of ECC6 is doing, integrating with the existing 3rd party components. That would be cheaper, solve the "going out of support" problem, and is less likely to crash and burn.
But looks less impressive on the Chancellor's CV.
>>> But looks less impressive on the Chancellor's CV.
No it won't. A University's Chancellor is typically a figure-head. You are probably thinking of the Vice-Chancellor who is more-or-less the University's Chief Executive. (Title may vary at different Universities but the distinction typically remains.)
In my day at Leeds the Chancellor was HRH The Dutchess of Kent, from whom I recieved my own degree diploma. Today it's Professor Dame Jane Francis.
Can't they get students to write such a system as a part of curriculum?
Universities should be able to prove that they can "produce" students being able to work in real world situations. Creating such system, that other universities could also contribute to could be one of the ways to refresh the image of stagnating education system.
Difficult, the complexities of such a system are beyond the time that can be dedicated to a student project. It's quite possible to graduate well from even one of the more prestigious institutions never having written a program over 1000 lines. Ideas of e.g. project management are essentially theoretical on projects of such limited scope. Requirements analysis for a project of this kind of scope would exceed what can be done in a third year project by itself.
Well, you have only got to look at the basic scenario which is a complicated and much modified system being "converted" into a SAP S/4 HANA system which has be plain vanilla.
Like other companies and councils already on this painful SAP route we are expecting to see lots of stories from Leeds with the headlines - delayed, over budget, review etc.
The UK's University of Leeds is looking for a systems integrator to help move its ERP system from an ageing SAP installation dating back to 1999 onto the German vendor's latest S/4HANA platform.
This is what happens when 'customisation' occurs. Why is it ever done...
It's done because some stupid bozos decided their way of working is better than everyone else's.
99.9% of the time the organisation needs to change their crappy processes which necessitated customisation in the first place.
As for the 11m, that would be my fee for touching such a god awful mess. However that doesn't leave a lot of wriggle room for the actual cost of the software.
This is so true, I remember working for a company whom customised the software for every single client so in the code there was plenty of this particularly egregious smell.
if (companyCode == "company")
Amongst other the olfactory offenders when you see a function 20,000 lines long split across 2 files it's time to run to the hills.
Remember if when being interviewed/sold too and you're told the software has grown organically politely end the interview/sales meeting there.
>>Why is it ever done...
Because every institution is different - whether it's a university, local authority, police service or whatever. You might think that something like a police constabulary could have a pretty standard business/process model, but talk to people from the Met and from Devon and Cornwall, and you'll find different. Local authorities are even more different - and universities strive to be distinct from each other, it's in their remit.
But this is not an argument for writing one system and nailing endless customisations onto it. It's an argument for institutions who are large enough to spend millions on SAP etc to employ their own IT staff and software developers, who are in daily contact with the institution's processes and people, and can put together a bespoke system which does AND ONLY DOES everything that institution needs. As a bonus, they get to enjoy job security by continuing to develop those systems in the light of the evolving requirements of the organisations. And the organisation gets the bonus that they are not paying for a massive amount of functionality that they will never use, but which makes the software they do use more fragile and harder to test.
Public sector services are not static systems whose entire workflow requirements for now and forever can be nailed down into a single specification and written into a contract - long experience clearly shows that this is a failed model. Instead of spending millions subscribing to cloudy products, and more millions getting contractors to customise those products - and most especially, instead of changing the established workflow of your organisation to match a one-size-fits-all product off the SAP/Oracle shelf - just spend fewer millions employing your own people who understand (and might even feel loyal to) your own organisation, to (continuously) develop a system which works how your organisation needs it to work.
And for those who will immediately cry "But what about all the time which will be wasted re-inventing the wheel???" - the wheel is being endlessly reinvented, in everything from F1 to inertial guidance systems to Swiss watches. Engineers know that wheels need to be round - but roundness is all that wheels have in common with each other*. All the other characteristics are optimised for the application, as good system design dictates - whether it's wheels or software.
*apart from Ferrari steering wheels, but they are just showing off.
You can tell I'm really old, can't you...
You just belong to a bygone era where organisations valued their staff, provided proper training, career paths and a job for life.
Hire a brilliant team of your own, write it all in house, maintain and modernise as you go is a fairytale. The reality is failing systems, unfit for purpose and impossible to maintain. In the real world, even the best designed and built systems fall into decay as the original authors move on, new staff fail to understand the system and new requirements bend the old designs past breaking point.
In 21st century IT, skilled and experienced staff are seen as a burden and cost centre, to be ruthlessly rooted out and cast on the bonfire. Ironically, often replaced by a bunch of cheap students.
Large organisations reap what they sow.
ECC 6.0 on Netweaver 7.5 is currently in standard support by SAP that ends 2027, when extended support is available through 2030. After that it's end of life. Dead.
So this article is two years out. That's more like the last dates to start a project given that 18 to 24 months would be typical.
Your milage may vary, and other support providers are available.
S/4 HANA Private Cloud Edition will allow Enhancement to standard if you really believe that the total cost of ownership is worth it..
The evidence is that more than 70% of custom code is not used so customers are deluded in believing that custom code is needed.
Clear leadership and governance are required as most customers are not in the software business...
Would you buy a car from your local higher education college because they run a mechanics course?
Professional software development is a highly skilled business that requires time and experience to learn. No matter how bright the students, they cannot deliver something like this.
A significant chunk of my business comes from companies that thought it could be done for free/on the cheap by a bunch of students. My experience is that 99.95% (approximately) of such code ends in the bin and 100% of such projects fail.
And that's before we open the can of worms about academics.
Would you buy a car from your local higher education college because they run a mechanics course?
People buy cars all the time from high schools with shop classes.
So yes. But as you prove yourself, you are comparing apples to oranges. A university not at the forefront of knowledge is useless.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2023
