Sheffield Uni cooks up classic IT disaster in £30m student project: Shifting scope, leadership changes, sunk cost fallacy

Sheffield University's failed Student Lifecycle Project went through three leaders, several changes in scope and was ultimately superseded by government policy change before the bulk of the £30m project was abandoned in what is shaping up to be a classic IT disaster. As The Register reported, the original plans for the Russell …

[Anonymous Coward]

So if I'm reading that right, they kept the obsolete system because they had no idea how to replace or upgrade the obsolete system. Then of course, nothing would TALK to the obsolete system because it IS obsolete and any support is going to be for CURRENT versions of the product, not something from the 90s.

This wasn't a project. This was some bean counter's glory hole for their ego, trying to stroke their own pride and power by "controlling" this huge disaster. I'm sure those three leaders had great power and fanfare before they were fired or gave up.

Yep. Nothing like watching bean counters try to drive technology. Without fail, in every single corporation or business I've worked for since 1988, the bean-counter driven projects have been abject failures because they're driven by "cost cutting" and vague "advantages" that have nothing to do with making a system actually WORK.

[Joe W]

I am (sort of...) involved in a bigger project of migrating critical infrastructure away from some very legacy architecture, developed about 50 years ago. Talking to some of the earlier developers you cannot even rely on the source code being correct, as some fixes were done in machine code in the compiled program, it was easier that way. There is a reasonable good understanding of what tries to talk to other systems, and how, but the project has been going on for a while already. The data exchange formats are well documented (mostly), the business requirements are very well written, but migrating this thing is a major effort. The project actually started by having everything new on more modern systems (cuneiform tablets would be "modern"!), forcing the new stuff to talk to the old stuff. Work is also migrated bit by bit from the central system, leaving the old thing in place for now.

Yes, it makes sense to start replacing the all encompassing system bit by bit, there is just no way it can be done in one giant project. Chosing a smaller subsystem as the first candidate while keeping the big, central data kraken in its place makes sense. There is just no conceivable way you will be able to replace that central abomination all in one go without a lot of money, a lot of work, and a lot of downtime. Pick any three of those.

(and an outdated Oracle DB is not the issue here...)

[cyberdemon]

> you cannot even rely on the source code being correct, as some fixes were done in machine code in the compiled program, it was easier that way.

Easier to ensure their long-term job-security that way?

Re. "Central Abomination" of R'lyeh:

At some point you would have to consider which would be more costly: An army of Oracle consultants incrementally "upgrading" the system, or A tragic 'accident' involving the core mainframe and its entire disk array.

[Tom 7]

At the time of the bodge its likely the programmer was respected and in a position to fix the bodge later, The bodge being something demanded by someone higher up. The the higher up shifts the goal past and outsources leaving the bodge in place for eternity despite warnings from the programmer. Its unlikely the programmer did it to keep their job, more likely someone higher up with no clue thought they could save some money, or gain power by outsourcing something that could actually be done far cheaper in house if left to the people that know.

I've actually worked in one place where programmers have, by refusing to tell contractors what their work involves or sign new contracts to make them do that. That involved a small council IT dept with 1/10th of the people in their IT dept comparted with similar sized councils that outsourced.

Some people take their jobs seriously and are best left to get on with it where possible.

[Fred Daggy]

Also, bodge now becomes standard operating procedure, because "Hey, it worked last time, right?"

[Stork]

Reminds me of a comment I once saw in a web app: “abominable hack:” roughly followed by “I did this and it appears to work, but I am not sure why”.

Here be dragons.

[Anonymous Coward]

"If it works, leave it alone!"

[Ken G]

And then if the requirements change?

[Cuddles]

It's abominable hacks all the way down.

[Anonymous Coward]

Actually, in the case of the Sperry AN/YUK-502 with the CMS compiler, you had no CHOICE but to correct the machine code because the compiler thought the results of a division were swapped in the registers. Unless you were writing code that never did division, of course.

Good old mil-spec. And we're supposed to trust such clown-like hardware with control of critical military infrastructure.

PS. The Sperry was still in use in the late 80's and early 90's because it had "radiation hard" iron-core memory and could withstand 50 caliber shells while being dropped 50 feet while running to simulate ocean waves.

Then there is the old VAX, which was kept "bug for bug compatible" to avoid breaking compilers on later generations of the hardware. History is rife with programmers having to work around buggy hardware and software through the most obscure methods.

Last but not least, my first *real* computer programming was done by POKEing machine code into memory and saving it to cassette. There WAS no code to compile, and not even assembly available.

[ricardian]

My Open University courses in computing (circa 1986) started by writing in machine code then using a simple assembler before moving on to OU Basic. Great way to learn - who remembers HEKTOR?

[Anonymous Coward]

HEKTOR? You were lucky to have HEKTOR. We had to made do with DESMOND,

(HEKTOR = Home Experiment Kit something something something, DESMOND = Digital Electronic System Made of Nifty Devices)

[Boris the Cockroach]

Desmond?

You should have signed up for a refresher course 12 years later.... dancing frogs(M206) followed by entering the dread portal of M301 and abandoning all hope as Bacon's masterwork* crushes the very soul from your body... although the first module was something rather pleasant involving java.. but only so you couldn't flee without losing all your money when Bacon came into view.

I still have no idea how I passed..... that year is a complete blank. (I suspect something to do with already knowing Java, and howto program multi-threaded applications for both computer animation and the real world.)

* Jean Bacon : Concurrent and distributed systems :An integrated approach (and not the wimpy edition she rewrote for the open university a couple of years later...... eeekekekkekk run for your lives..!!! )

[Antron Argaiv]

I once had a mangler who told me to do that...on an FPGA. "Just go into the placement editor and fix it"

I told him I wouldn't do that because it made the sources obsolete. He didn't see the problem. I did, and it wasn't the hardware or the code...

Got out from under him ASAP.

[David 132]

At some point you would have to consider which would be more costly: An army of Oracle consultants incrementally "upgrading" the system, or A tragic 'accident' involving the core mainframe and its entire disk array.

I've been reading too many BOFH columns because my first thought was, "a tragic accident whereby the core mainframe and entire disc array fell out of the 20th storey window onto an army of Oracle consultants".

[Anonymous Coward]

> So if I'm reading that right, they kept the obsolete system because they had no idea how to replace or upgrade the obsolete system.

Worse, they then decided to make it a hub without fully understanding what it does. A textbook example of how skipping documentation and the boring hygiene stuff costs many, many times more when it has to be done retrospectively. (Or in this case, not done.)

I wonder if Sheffield Uni has a business school where they could teach this as a case study?

[Pseudonymous Clown Art]

To be fair, its not *just* beancounters...it's also the "why upgrade? it works just fine" brigade.

There is also a massive problem in IT with familiarity...it trumps functionality every time.

[Stuart Castle]

Ahh bean counters.. A few years ago, as part of my job, I supported a small TV studio for student use. This studio would have been state of the art, except the bean counters got involved. They slashed the budget half way through installation, with the result that we had a room that was designed from the ground up to be a studio. It had excellent sound proofing, a large green screen area, and curtains heavy enough to block out 90% of sound, with the walls being designed to block out 100% of sound from outside the room. The curtains were designed so we could curtain off sections of the room with near total sound insulation, and state of the art lighting. It also had tons of storage for the studio equipment, as well as full swipe card entry.

The one thing it didn't have was a control room. So, we had no control over the lights, beyond a small control box the installer had left so we could actually use the installation. We also had no mixers, beyond a couple of small ones we had to lend to students to use in the field. We were promised state of the art video and sound mixers, with "talkback" (where the control room staff can talk to the cast/crew), and a state of the art fully integrated lighting control system with computer control.

[Alan Brown]

"The one thing it didn't have was a control room. So, we had no control over the lights, beyond a small control box the installer had left so we could actually use the installation. We also had no mixers, beyond a couple of small ones we had to lend to students to use in the field."

Oh dear, the control box has caught fire and the loan mixers have come back broken. We can't use the studio and it's now an expensive white elephant. Whatever would the Public Accounts Committee say?

[Jellied Eel]

PAC you say? Possibly wondering why another small studio commissioned a Banksy artwork so Christopher Walken could paint over it. Ah, the public sector.

[TRT]

We had a studio and control room / gallery and voice-over booths etc.

It caught fire (well, something outside caught fire and it spread.

We were asked by the insurers to cost a replacement.

The original studio and control room etc had actually been cobbled together by a succession of ex-BBC engineers who had retired and got bored in their early retirement so decided to get part-time jobs at the local technical college.

The cost for replacing it with comparable commercial kit was over 5 million.

Their assessor hadn't done a very good job in setting the premium! Still, they paid up, I'll give them that, and we had a first rate training facility for all of 2 years until the manglement fuzzed up the financial position so badly that the college effectively closed. They stopped offering any courses even remotely useful (they tend to be costlier to teach), sold the building (including the studio), moved into smaller accommodation and from then on it was practically all business studies, management courses and social sciences instead of printing, media production, music production, engineering, photography, ceramics etc.

[unimaginative]

"But it wasn't really sold to the business in those terms because it's not sexy."

That demonstrates part of the problem with universities. They are supposed to be non profits, but act like businesses.

Not doing stuff because its "not sexy" is a more widespread problem, of course.

[Anonymous Coward]

They are supposed to be non profits, but act like businesses

Sorry to bring bad news, but Universities are businesses. They have marketing departments, they have recruitment targets, grade targets, research income targets, spin-off income targets. Spending is examined on whether it will help the University's targets.

Universities don't just survive on government education grants. Research, commercial teaching (Have you seen the price of an MBA course...?), etc.

I've heard (sorry, no hard evidence) that some UK universities get so much of their income from outside/commercial sources that, technically, they are no longer UK public sector.

[unimaginative]

That is exactly the problem.

They are legally not for profit organisations, but engage in business. In terms of efficeincy, corporate governance and being generally well run that is the worst of both worlds.

Legally (most?) universites are not public sector bodies. They are perfectly free to walk away from the government funding and raise their own. A number of Russel group universities did threaten to do that just a few years ago if they were not given more funding or allowed to raise tuition fees.

[Anonymous Coward]

They are legally not for profit organisations, but engage in business.

Well of course they do. My own beloved employers (a university with charitable status) have an annual turnover well over £500m. An organisation of that size has to be run on businesslike lines, as for that matter does the East Harptree Women's Institute.

[Tom 7]

TBF MBA courses get to charge what the idiots will pay to join the club. They dont actually seem to have any genuine expenses from what I've seen of their output.

[Jim Whitaker]

Public Sector - I don't think so.

Real Universities never were "Public Sector" and nor should they be now.

[Anonymous Coward]

Re: Public Sector - I don't think so.

Polytechnics were owned by LEAs, as FE colleges are now.

[A Non e-mouse]

"But it wasn't really sold to the business in those terms because it's not sexy."

That's bad project governance.

[Ian Johnston]

They are supposed to be non profits, but act like businesses.

Of course. "Non-profit" is a subset of "business", not a mutually exclusive concept.

[ArrZarr]

The problem with any concept that you can configure something without code is that you always need something just as complicated as code to actually do the heavy lifting in the configuration space.

People keep falling for this concept without realising that anybody who can understand the labyrinthine rules behind the codeless configuration will be fully capable of dealing with code based configuration anyway. Do the managers really think that it's (a) going to be possible for them to futz around in the config without a huge learning curve or (b) anywhere near a good idea for somebody who hasn't climbed that curve to mess around in a configuration file?

[Valeyard]

30 million bloody quid, easy to piss money away when you can hike up the student fees

[Chris G]

Piss Poor Planning

Provokes pissing pennies up the wall.

It sounds as though the Uni' needed to make a plan of how to make a plan to replace the old plan.

[Lazlo Woodbine]

They can't hike fees though, as they're capped...

[Valeyard]

"capped" means they're already raised as high as they'll go

[Alan Brown]

Except that these days students are demanding valie for money and starting to take legal action when they don't get what they paid for

Universities are frequently the last bastion of British Leyland manglement

[Pascal Monett]

What this boils down to

is that the uni spent £30 million to learn that they had to replace everything else before implementing a Student Lifecycle Project, which is just fancy wording for an SQL database of student grades.

Oh well, sometimes lessons are expensive.

I would have said expensive lessons are the best remembered, but somehow I doubt that the uni is going to remember anything from this.

[Doctor Syntax]

Re: What this boils down to

Expensive lessons are likely to be buried in unmarked graves.

[batfink]

Re: What this boils down to

Quite. We laugh when we read "£30M for a Student Record System".

However, working out WTF the current systems do should have been an entirely separate project, not tied into this one. Then once they'd done that, they could have spent peanuts on the Student Record System.

[Anonymous Coward]

Re: What this boils down to

Bean counters stay focused on the actual SCOPE of the original project? Heaven forbid when there are kitchen sink wishlists and wants to add to the scope creep until a $250,000 student record project by a couple of grad students gets morphed into re-engineering the entire set of business systems to the tune of millions.

What baffles me is that the bean counters who signed off on the project haven't been fired. Sure some people got the ax, but they were just running the project, they weren't the ones who signed off on it. Usually when you're talking this kind of budget, someone(s) at the A or B level of management gets a very public axe and all the blame...

[Alan Brown]

Re: What this boils down to

"What baffles me is that the bean counters who signed off on the project haven't been fired."

They get peerages, a seat in the House of Lords and EVEN MORE contracts

[Anonymous Coward]

Re: What this boils down to

I know right.

[Anonymous Coward]

Re: What this boils down to

How do people find these gigs? I never get anything as choice as this. £30m to tell someone their system sucks is an awesome gig.

I'd like to know for a friend.

[cantankerous swineherd]

"data models between SLP and CIS were fundamentally incompatible"

here's me thinking that munging data from one form to another was what programmers did.

[DJV]

I expect there was a lot of "if you'd told me up front you wanted to do THAT with the data then I wouldn't have programmed it THIS way in the first place."

[Doctor Syntax]

I've been in a situation where I'd rather have been given what they programmed in the first place. What they (client's client) extracted appeared to have had several fields concatenated into one. They had to be taken apart again. I even had to get them to add a flag to tell me just what it was they'd done.

[DJV]

Concatenated fields

Similar story except, in my case, the client "didn't want to bother" me by asking for a new fields in a customer database table, added secondary data (I think it was eBay and/or Amazon IDs and sometimes both) to the phone number field. 3 years down the line they needed to have this data separated and it took far longer (and therefore cost them more) to unpick the mess, decode which bits were phone number(s) and which were IDs and shove them into separate fields.

[Anonymous Coward]

here's me thinking that munging data from one form to another was what programmers did.

This isn't just a case of converting between, say, CSV & XML formats. It's going to be the data in their CIS can't be converted into the HESA reporting format.

At our place we were held back in amending the gender field in our systems from a simple Male/Female into a broader ranger covering LGBTQ+ because the upstream systems we fed into couldn't handle it. It'll be a similar thing a Sheffield. (But on a bigger scale)

[eldakka]

Still doesn't make a lot of sense.

This is what ETLs, brokers, ESBs, and portals are for.

[eldakka]

> This is what ETLs, brokers, ESBs, and portals are for.

A broker could absolutely handle the use-case proposed:

>> At our place we were held back in amending the gender field in our systems from a simple Male/Female into a broader ranger covering LGBTQ+ because the upstream systems we fed into couldn't handle it.

Upstream systems could be fed through broker interface X that identifies the receiving system and parses the gender field retrieved from the system and 'translates' it depending on the raw gender LGBTQ+ inclusive value to a value acceptable by the upstream system (NULL or Male/Female). If the upstream system is identified as one that is complaint with the new gender format, it passes it through unchanged. This is what brokers do.

[Anonymous Coward]

A true data integrator has spoken

Interesting. What is that 'raw gender' you talk about and how would you suggest to get that data?

Let's have as an example LGBTQ+ gender value 'Transgender'. Male? Female? Please do enlighten us.

[Dr Scrum Master]

At our place we were held back in amending the gender field in our systems from a simple Male/Female

Sounds like a good thing

[Anonymous Coward]

Generally, yes, but I find it really helps if the people in charge of, say, the SLP data models can actually tell you in under three years what the hell those models are supposed to look like and contain.

We eventually gave up asking them for sample data and just watched them repeatedly bulldozing money into a flaming pit hoping that *this* pile of notes would be the one that would finally put the fire out.

[GreyWolf]

Not so fast, my young poadawan

"here's me thinking that munging data from one form to another was what programmers did."

That only works when the item represented is the same thing in both cases.

There are lots of instances where it's only the Hoomans that use the same word for two different data objects.

I worked for a building society where the "house" that you had a mortgage on was not the same object as the "house" you had an insurance on.

[elsergiovolador]

Contractors

and we started to have a lot of churn of those contractors, so they acquired a lot of knowledge and then they would leave

That usually happens when the complexity and scope turns out to be not adequate to the money on offer.

[Doctor Syntax]

Re: Contractors

It can also happen when contract admin forgets to renew the contracts or there's a freeze on spending.

[Anonymous Coward]

The fundamental here is that they tried to modernise a legacy set of systems by implementing a new system. Rather than take a brown field approach cementing existing business processes structured around the legacy systems they should have taken a green field approach and implement a cloud student and ERP system and changed the business processes to fit with best practice blueprints for England HE. Every English University has exactly the same reporting and record keeping needs as they are set by government, you don’t need to analyse them to death or design your own bespoke processes to manage them. Honestly, you could have a three ring binder on how to operate a English University and it would 99% work for all universities.

[Bitsminer]

But inevitably someone would try to automate that 3-ring binder. And another 30million later it would still not work correctly.

[Jellied Eel]

First define your 3 ring binder. Landscape or portrait? Hole in each corner, with additional costs to format and print on triangular paper.

[Alan Brown]

> Every English University has exactly the same reporting and record keeping needs as they are set by government, you don’t need to analyse them to death or design your own bespoke processes to manage them.

"But how can we justify our own jobs if we don't make it entirely baroque and bespoke?"

This is pretty much the case. I have had a lot of struggle trying to bring in off-the-shelf systems over "we can write this here" which ends up not doing what's required and costs 10 times more

[Anonymous Coward]

Unfortunately "off-the-shelf" systems are usually anything but. They tend to customise them to fit customer demands and they end up as Frankestein's monster. Only worse.

For example, and to be honest this is something a lot of function libraries suffer from and don't take into account, we have numerous "year spaces". There's an Academic Year, which follows term dates (variable), a UK Tax Year (easy one), the Calendar Year (easy again), the Financial Year (slightly harder - it's offset from the academic year by -1 month), the Academic Financial Year (offset from the academic year by +1 months), and the Financial Reporting Year (as the university is now a business, this and subsequent reporting quarters are tied in to the Financial Reporting Date as submitted to Companies House). Complicating this further is the fact that several internal functions were moved into single-customer incorporated service businesses which have their own set of calendars.

So, when generating a financial report, say, the date range field is filled in, for example 202000 to 202012. But a part year report is filled in say 202008 to 202012. This doesn't mean August to December, it means the start of the 8th month of the reporting year to the end of the 12th month of the reporting year, and the reporting year changes according to which account code is being quizzed - is it a teaching account which follows the AFY or an internal income account which follows the FY? On top of which semantically 202000 and 202001 are the same - start of the first month of the period and start of the period absolute, which means are there 12 or 13 months in that period? Nowhere is there a drop down menu. Nowhere is the field pre-populated with and indication of which calendar is being referred to. And if you use it infrequently and forget these foibles, nowhere does it indicate the actual reporting dates covered - it just repeats the same date code as you entered, so you could forget the 08 at the end of a year (and does it mean 2020/2021 or 2019/2020?) refers to the eight month and not August. And this is the NEW system that's just been signed off! It's a monster! We paid for it.

[TRT]

Financial Quarters

I hear you. The MySQL Quarter function has to be one of the most useless functions around. It returns 1 if the month is between January and March inclusive, 2 for April to June etc...

BUT businesses can select their own accounting reporting period, so Q1 for Apple starts on Sept 26th I think it is, whereas Oracle's is June 1st.

I did suggest an extension to the function whereby the Fiscal Year start date could be passed to it, but for some reason I got bawled out of the forum by a couple of irate developers.

[saif]

Inevitable

After decades of evolution of systems inevitably develops capabilities that can not easily be replicated in a modern, up-to-date , monolithic project, even as the legacy technology becomes obsolete. As any one associated with NHS IT knows, revolution always leads to (hopefully temporary) degradation of performance at greater expense than just evolution...outsourced contractors just have to make promises, charge big bucks, then complain the problem is with the legacy system, and finally deliver an inferior product. This product, without benefit of bedding in and evolving itself, is thrown away because it cant compete with decades of adaption to user needs.

[andy 103]

This is why the world runs off spreadsheets

You're building all these views and processes and you get all of the problems that you would do with conventional software development, [however] you just don't have any of the standard tooling that makes your life easier while you're doing it.

So much I could write about this. Most applications used by large organisations are simply unfit for purpose. The reason that so much data is siloed in spreadsheets and sent round by email is because - shockingly in 2021 - this is still one of the most effective ways to get data in front of people that need to see it. This then introduces a massive problem because there is no single source of truth. People assume what they've been sent or have in their Inbox is always "correct".

When you have terabytes of data it goes without saying that the majority of people who use that data only need to see / work with a portion of it at any given time. A lot of these systems are designed without any consideration of end-users in mind. So when they need to get that data in front of them they simply export portions of it that are needed (even if that export involves manually copying/pasting data). They then manipulate it using Excel or some other such tool and send it on via email because, well, how else when the system itself doesn't even produce the reporting people need.

the old system could not support the new data standards, but that was never promoted to management.

Even if it was, they wouldn't care or even understand what this means. As long as it's presented as a financial figure, the lowest value will always win.

But as work began developers realised it was connected to so many other legacy systems – with little or no documentation – that this was not going to work

Documentation and handover costs money. See previous point.

This happens in _so many_ organisations it's frightening.

[elsergiovolador]

Re: This is why the world runs off spreadsheets

Once every couple of years there is new energetic management with a bright idea to get people off of their spreadsheets.

They mandate workers to turn over their spreadsheets they use for day to day work. The analysis is commissioned, user research, requirements gathering, interviews. Spreadsheets are turned into specifications, contracts are written, contractors appointed. Finally this time it will be the end of it.

By the time the work is completed, the workers already use completely different spreadsheets and freshly built system largely makes no use for them.

They end up using their spreadsheets and then also entering whatever they can to the new system (and the old system from previous management).

Then someone comes with an idea - let's just create a spreadsheet online. Everyone laughs and cries.

Rinse and repeat.

[Ken G]

"let's just create a spreadsheet online"

Put it all in MS Access!

[TRT]

Re: "let's just create a spreadsheet online"

Or a less proprietary data management system!

[ecofeco]

Re: This is why the world runs off spreadsheets

Perfect comment is perfect.

[Dwarf]

If they have such a complex system that not many understand, then it would be a really good idea to document it at the various levels using any recognised architectural process (TOGAF etc) so that at least they then know how big the problem is and what happens if you pull this piece of string or change that interface file.

Just stopping a programme doesn't make the lack of knowledge about their current systems go away, it just buries it for a bit longer so that it can rot a bit more.

[Anonymous Coward]

Some benefits though

Yes, at least they came out of this project having carefully examined and documented those complex interfaces so they've got some step forward to show for it when they next try to tackle it.

Surely.

Definitely.

Anything else would be the utmost foolishness.

[David Lewis 2]

Put a positive spin on it.

At least the university has a new case study they can use for future courses.

[Anonymous Coward]

WTF

So their present IT director has no IT background. did any of the previous ones know anything about IT?

Explains a lot.

BA in history and MBA!!!! fucking clowns

[Tom 7]

Re: WTF

My OCD gets triggered by people with no knowledge of IT managing IT projects - and this frequently includes the customer. Quite a few times I've written code that does what I believe the customer actually wants rather than what the consultant and customer have come up with together and then found some excuse to show it working to them much to their great delight.

Working for an MBA is like getting in a taxi with a driver who cant drive, doesnt know where you are, doesnt understand one way streets or road maps and hasnt a fucking clue where your destination is, its often quicker to walk.

[TRT]

Re: WTF

Management is made up of transferrable skills.

I don't need to be Geoff Boycott in order to deliver an effective whack to the back of someone's head with a cricket racquet!

[Anonymous Coward]

Re: WTF

Degrees, etc != experience though. Really triggers me (as someone who has successfully worked in multiple functions in IT for over a decade with no degree in IT) when people ignore that professional qualification, real world experience, and Higher Education in ANY subject gives ample skills for working in IT and knowing how to run projects. IT is not a mystery and most things can be learned by almost anyone quite quickly. I have rarely met people working in IT roles with IT degrees, it's normally anything from Dance & Arts to Zoology, or shock, no degree at all! A quick Google suggests that the current IT director has approx 20 years experience in different IT functions and at senior levels in the Education IT sector and is recognised as a board member of an industry body. What more would someone have to do to be worthy in your eyes?

[Anonymous Coward]

Re: WTF

What you mean is they have talked bollocks to upper management for 20years and enjoyed being wined and dined by marketing/sales wankers.

Like most upper management IT fuckwits who actually know nothing about the underlying technolodgy other than what marketeers have told them when being wined and dined (more bollocks).

Surely the number of failed IT projects ran by MBA, dance and other fuckwits would be a clue that it's a fucking terrible idea.

Being on an industry body just means your good at wining and dining with fellow fuckwits.

Only time they accidentally succeed is that they had an underling WHO had proper knowledge. and was actually intelligent enough to fix the messes the fuckwits make, nothing to do with degrees either (But actively getting an MBA proves your a fuckwit to start with!) some of the best tech people I know don't have degrees, But they know more,can fix and design better than any degree person who only goes by the stupid bollocks they were taught.

[Potty Professor]

Re: WTF

Couldn't agree more! My degrees were in Mechanical Engineering and Diesel Engine Technology, but I worked for many years in an Electrical Engineering company, rising to the dizzy heights of Principal Engineer, Technical Manuals Department. (I was offered a Management position, which I turned down in favour of the PE position, but that's a different story). I then moved to a different company, in Publishing, and became their Deputy IT Manager in the late 90s, a position for which I had absolutely no paper qualifications, but for which I had oodles of experience (obtained both at the EE company and the Publishing company), and went on to oversee the relocation of Head Office to a new location, with all new IT infrastructure.

[Anonymous Coward]

"That required consultants."

I didn't need to read any more - those three words said it all.

(I did continue reading of course, this is The Register after all).

[Anonymous Coward]

I'd be tempted

To advise people to go somewhere else for Comp. Sci degrees

[TRT]

Re: I'd be tempted

Nah. You're talking about Sheffield here. Their reputation is stainless.

[Anonymous Coward]

I am not surprised one bit. Legacy systems work, but they are reaching end of life. New systems do not work. Bureaucracy expands. The services departments expands and centralize. Centralized services do not understand what is happening in the departments, so require more data. Bureaucracy expands. Costs increase. The data provided by overworked persons at the department is sometimes fake, as nobody cares and it is easier to fake than to monitor. The new centralized implementing 'strategies' loosely based on the available data and more on wishful thinking do not work. Clearly that is because there is not enough data fed to the centre. Bureaucracy expands. Costs increase. The same people at the department are asked to feed more data into new systems. Data is clearly fake, as it is easier, and it is the only way to show compliance with the strategy of the institution. And so it goes, with occasional outsourcing of something...

[DomDF]

Blind leading the blind...

straight through a field of nuclear landmines.

[GreyWolf]

Been there, done that....

I have seen this happen so many times in building societies, banks, insurance companies, etc etc etc.

You face an impossible dilemma - do you (1) rewrite it all simultaneously (boil the ocean) or (2) write something new and feed it from the old (where none of the data is compatible*) or (3) buy a software solution from USA? All routes lead to disaster.

* Incompatibility: what you mortgage is not what you insure, and ne'er the twain shall meet. None of the data fields match.

The companies I saw trying to solve this riddle screwed themselves up, are now merely brand names, and no longer exist in any form.

Except for one, which is going to solve the problem by getting rid of not just the software, but also the entire product range and all the customers, 26 million of us here in UK are going to be just dumped. And left without an emergency service.

[mpi]

Out of curiosity, here is a question:

How much would it have cost to rewrite the system from the ground up?

[Antron Argaiv]

Well, first you need to write a requirements doc...

[Tom 7]

No, first you need to actually know what the old thing does. Then you can map that to a requirements doc. The current users may think they know what it does for them but most likely dont.

[yetanotheraoc]

Parachute not optional

"How much would it have cost to rewrite the system from the ground up?"

That's what they are going to find out as they replace the out-of-support Oracle system.

From the article. "At Sheffield University, a new director of IT services, Bella Abrams, joined in 2019. **Following** (emphasis added) her arrival, it was realised that one of the main reasons for doing the project in the first place – statutory reporting – had become unnecessary."

Likely it was realised well before her hiring. But first, the previous IT manager needed to engineer an escape. Now they need a new new system, and it will be another boondoggle. The only question is whether Ms. Abrams will receive the same favoured treatment.

"Then we got in a lot of contractors and we started to have a lot of churn of those contractors, so **they** (emphasis added) acquired a lot of knowledge and then they would leave," the insider added.

If only they had gotten a complete description of the current system for their £30m, but I bet they didn't even get that. Because that would have been one of those hard efforts they so carefully avoided.

[Chris Coles]

Out of curiosity, here is a question:?

Thank you, the best answer herein has to be the curious question . . . why not set out to start from scratch and create their own system that then could be sold to every other university?

[claimed]

Re: Out of curiosity, here is a question:?

Uh-huh

https://xkcd.com/927/

[TRT]

Re: Out of curiosity, here is a question:?

Great idea! Every university should do it!

[The Basis of everything is...]

Shouldn't a University have known better?

A few decades ago (ouch, writing that hurts) when I was doing my time at a polytechnic having been deemed by the likes of Sheffield to be not worthy of getting a proper education, this sort of thing was featured in at least two of the courses as classic how not to do it.

So if the poly's were teaching this stuff to the undeserving masses back than, and Sheffield is supposed to be stuffed to the rafters with brains the size of a planet, just what has gone wrong with our educational elite?

[Anonymous Coward]

Re: Shouldn't a University have known better?

It's almost unknown for universities to consult their own academics on matters of institutional policy within their areas of expertise.

[Anonymous Coward]

Re: Shouldn't a University have known better?

Even when the academics are ex-industry with 30 years experience. I've seen it both in IT and building/planning.

[Anonymous Coward]

Re: Shouldn't a University have known better?

The ONLY instance of this happening that I can think of is over the last two years. Mind you the academics were generally the very same epidemiologists that the government were asking. But the government also threw out the bone that workplaces were responsible for their own risk assessments and control measures... and you couldn't get a consultant in for that because (1) time and (2) availability. So they were effectively forced to think inside the box for once.

[ecofeco]

So much money

... for so MUCH stupid!

Good work if you can get it, I guess. Meanwhile, the rest of us struggle to get paid for decades of knowledge and proven results.

[adam 40]

£30M on a student project?

How times have changed, in the '80's in my final year (Computing/Electronics at Durham) I couldn't get the department to buy a TMS320 dev board for my adaptive filter project. Cheapskates!

[EricB123]

Once Again, My Words of Wisdom

The 7 Phases of Project Design

1) Wild enthusiasm

2) Disillusionment

3) Chaos

4) Search for the guilty

5) Punishment of the innocent

6) Promotion of the non-participants

7) Definition of the requirements

This happens almost every time!

[TRT]

Re: Once Again, My Words of Wisdom

How do I bookmark this?

[ExImperialITperson]

£30 million down the drain that is nothing - Imperial college's spent £50 million over budget on their student system.

They also have an Oracle financials that is going out of support in a few weeks and last year replaced the experts who had worked on it for years because of senior management feeling threatened by them.

[Anonymous Coward]

just checking the Oracle version on oor production servers...

10g 10.2.0.4.0

Our old IT director (retired a few years back) in his infinite wisdom negotiated a kind of "perpetual, no upgrades" license. Apparently. This stiff went out of support more than a decade ago. The thinking was always "it's internal systems, behind firewall". Now we have production sites inside the PRC and have to go through contortions to keep the CCP out of our systems...

Aah fun and games.

[Jim Whitaker]

Previous failed attempts

Anyone else on here remember a project called Delphic? So much of what is reported in the article brings back horrible memories. And a smile or two.

[Anonymous Coward]

Re: Previous failed attempts

The core students records system at Sheffield was begat by the Delphic project, I dare say the backend databases still have MAC in their instance names which gives another clue to its origins. On the point of it being out of support from what I recall of the state of the system in 2016 I see no reason why the underlying database and middleware couldn't be upgraded to a supported level but I guess the intention was to replace it so it was left to slowly go obsolete. It's funny how often this story is repeated with legacy systems still hanging around as the replacement projects never finished or never manage to replace all the functionality of the old system it was replacing so you just end up with an old and a new system to maintain forever.

[TeeCee]

Aha! Found the problem.

That required consultants.

Wurp, wurp! Pull Up!

Wurp wurp! Massive cost overrun approaching!

As soon as somebody decides that getting in consultants is a good idea, the wheels are off and it's time to pull the eject lever. The alternative is moving the contents of your bank account into the consultancy's bank account.

[adam payne]

A Sheffield University spokesperson said: "The investment in the programme has already delivered new systems that previously did not exist at the university and our work to date will be important as we develop our future plans."

Yeah of course new systems were delivered, they don't work but it sounds good on a statement.

[Anonymous Coward]

Scams

What I learned was that these car crashes aren't normally the product of inescapable complexity or incompetence.

Normally it's because there's a big grift that can be run while the thing is failing. My guess here is student loans, I would hazard that the bad systems and chaos caused by the failed replacement are being used to extract loans from fictional students which are then stolen. I would guess that a hierarchy of kickbacks and threats is being used to keep the disaster rolling so as to keep the door open to the fraud.

[Anonymous Coward]

I love it when IT management know nothing about IT as such

Many moons ago I was in a meeting (me being a lowly programmer fresh faced from my student days) and the IT mangement who were all ex engineers who had migrated into the university's IT dept.

One of them had the BRILLIANT idea that since our RAID hdd's hadn't had a single fault in the past 4 years, why can't we just un-RAID them and thus double (or more) our storage capacity with no extra cost.

Lucky there were a few other people who were clued up enough to shoot the idea down (with nukes). But the fact that the idea was brought up made my blood run cold....