What is it with government and IT projects?

[Deleted member 66701]

Then your bosses were fools for allowing DWP to push SLAs without increasing the fee.

No arguments from me there! One of the reasons why I'm no longer with them.

 

[McGraw]

Can't help but think of iSOFT as I read this.

 

[elmarko]

frankly if you are in charge of a project that fails due to mismanagement (rather than say the company you contracted going bust etc) then and wastes hundreds of millions but sorry you should be fired for incompetence/negligence.

yet in the public sector it seems you just carry on moving up the seniority ladder of promotion.

Having being involved with many large IT projects in the private sector I can personally say this is not a problem specific to any one area (public/private).

It's to do with a lack of technical expertise within the higher echelons on management in most organisations which seems to be the larger issue (as they can neither verify or validate the claims of a third party - or know which internal 'experts' have a valid opinion).

There are also the other reasons which have been very well explained in previous posts.

 

[TommyV]

I never understood how a company can bid for a contract and then it ends up costing three times that

Companies will push their OOTB product, it's bespoke functionality on a customer by customer basis where most of the money is made.

Then, moving forward, they'll have that bespoke functionality in their back pocket which they can sell to other customers and charge hefty premium for it.

 

[quickshot89]

HP have lost other Gov contracts due to general incompetance, so it doesn't surprise me if they lose the DWP too. EDIT: DWP seems to be moving to a more SIAM like model, so HP will certainly lose their current contract, and have to settle for a smaller slice of pie, or lose out completely.

Has anyone got any links to the failed projects they are posting about?

it would depend on which part of the contract your referring to, hosting, apps, desktop etc are all different

 

[dirtychinchilla]

My dad has worked as a PM for Logica and now CGI for about 25 years, and he says constantly that no one ever budgets enough money for anything...reflects what most people in here are saying.

 

[anything I don't mind]

I think it stems from the nature of public sector itself. They don't have the same threat of bankruptcy. It is far too centralised and to try and approach IT systems with centralised systems like that must be difficult. I think it would make more sense to build each node up first and then slowly link them all up. I don't see why it would be difficult to implement an IT system for health records. Definitely not £12 billion. I think 100th the cost of that would be more reasonable for a project to link doctors together for digital health records. Could even do it all remotely with one database server. lol. All they need is internet in each GP and a login password and its all set. Minimum cost with high quality probably about £100k. Assuming each GP already has a computer with internet. I think it must just be digging holes and filling them mentality. People just doing nothing all day and getting paid £millions.

 

[Judgeneo]

I love how in one paragraph you have gone from £12B, to £12M, to £100k. Have you considered a career as a salesman in an Gov outsourced IT provider?

 

[Dj_Jestar]

There's also the part about keeping the entirety of the NHS data on a single server, where "all they need is internet in each GP and a login password and its all set"

We should frame that.

I wonder if he realises that that budget also includes the hardware?

 

[elmarko]

I think it stems from the nature of public sector itself. They don't have the same threat of bankruptcy. It is far too centralised and to try and approach IT systems with centralised systems like that must be difficult. I think it would make more sense to build each node up first and then slowly link them all up. I don't see why it would be difficult to implement an IT system for health records. Definitely not £12 billion. I think 100th the cost of that would be more reasonable for a project to link doctors together for digital health records. Could even do it all remotely with one database server. lol. All they need is internet in each GP and a login password and its all set. Minimum cost with high quality probably about £100k. Assuming each GP already has a computer with internet. I think it must just be digging holes and filling them mentality. People just doing nothing all day and getting paid £millions.

Have you ever worked in the private sector, the public sector or in a role related at all to large scale IT projects?.

This has nothing to do with it being public or private, it's simply the mechanics of unrealistically low estimates winning contracts & the mechanics mentioned above.

Building separate nodes would be an utter disaster (unless they all built to the exact specifications & linked, but if that's the case when what's the advantage of building them split at all?) as attempting to merge countless sporadic systems into a single system at the end is the most difficult tasks imaginable for a large scale IT project.

I love how in one paragraph you have gone from £12B, to £12M, to £100k. Have you considered a career as a salesman in an Gov outsourced IT provider?

There's also the part about keeping the entirety of the NHS data on a single server, where "all they need is internet in each GP and a login password and its all set"

We should frame that.

 

[anything I don't mind]

No I mean by nodes each GP. Make each GP have computer and internet and then when its all ready build a central server with database and have the GP run a vpn client then open a web page to log their patient records. The difficult part would be getting a way to import old and existing records in to the system. The easy part would be setting up a system to log the patient records. It would not near cost £12 billion. I don't even think i could think of enough reasons to get a project like that to £1 billion even if i tried the most expensive options for everything.

You could probably mpls link every GP in the country for less than £12 billion. Well you could definitely upgrade large sections of infrastructure for the amount. Like fiber to every home for example.

It has everything to do with the public sector. I have never heard of a £12 billion waste of an IT project in the private sector. That just does not happen. People would be fired long before it got to that point in the private sector. Plus if it was private sector they wouldn't be wanting to have a centralised system as they want. Each GP would be independent and just share the records upon request.

 

[Tunney]

No I mean by nodes each GP. Make each GP have computer and internet and then when its all ready build a central server with database and have the GP run a vpn client then open a web page to log their patient records. The difficult part would be getting a way to import old and existing records in to the system. The easy part would be setting up a system to log the patient records. It would not near cost £12 billion. I don't even think i could think of enough reasons to get a project like that to £1 billion even if i tried the most expensive options for everything.

You could probably mpls link every GP in the country for less than £12 billion. Well you could definitely upgrade large sections of infrastructure for the amount. Like fiber to every home for example.

 

[Dj_Jestar]

According to this article from 2014 there are over 32,000 GPs in the UK (32,075 is the number in the article).

What are you proposing to supply them with for £3.12p (and a half) each?

 

[anything I don't mind]

According to this article from 2014 there are over 32,000 GPs in the UK (32,075 is the number in the article).

What are you proposing to supply them with for £3.12p (and a half) each?

It would be £100k up front and then a monthly payment for the internet connection. Set up a central server, collocate at a datacenter for example with 1gbit link. Then you use existing web based medical records software instead of developing bespoke. The software would be at an additional cost as i have no idea how much they cost i estimate about £30-50k but maybe more. Set up a firewall vpn that can take that many connections. The additional cost would be digitising the old medical records and importing them in to the system, that is not included in the £100k.

 

[Abraham]

No I mean by nodes each GP. Make each GP have computer and internet and then when its all ready build a central server with database and have the GP run a vpn client then open a web page to log their patient records. The difficult part would be getting a way to import old and existing records in to the system. The easy part would be setting up a system to log the patient records. It would not near cost £12 billion. I don't even think i could think of enough reasons to get a project like that to £1 billion even if i tried the most expensive options for everything.

You could probably mpls link every GP in the country for less than £12 billion. Well you could definitely upgrade large sections of infrastructure for the amount. Like fiber to every home for example.

It has everything to do with the public sector. I have never heard of a £12 billion waste of an IT project in the private sector. That just does not happen. People would be fired long before it got to that point in the private sector. Plus if it was private sector they wouldn't be wanting to have a centralised system as they want. Each GP would be independent and just share the records upon request.

I know you're just using it as an example, but GP systems are in relatively good shape. Over 50% of GPs use emis Web, which is fairly poorly received but provides a linked up system with a level of patient access, a large percentage of the rest use either of the big two competitors, who offer a better product, again providing the same functionality.

The biggest problem within healthcare IT is communication, between hospital departments, out to community and back to primary care (GPs). People die because of miscommunications (by people) at these touch points, yet despite government targets encouraging it, improving it is fairly low down on most tenders we receive.

 

[anything I don't mind]

I know you're just using it as an example, but GP systems are in relatively good shape. Over 50% of GPs use emis Web, which is fairly poorly received but provides a linked up system with a level of patient access, a large percentage of the rest use either of the big two competitors, who offer a better product, again providing the same functionality.

The biggest problem within healthcare IT is communication, between hospital departments, out to community and back to primary care (GPs). People die because of miscommunications (by people) at these touch points, yet despite government targets encouraging it, improving it is fairly low down on most tenders we receive.

So they spent £12 billion on an IT project when they were already using a web based third party? Where did the £12 billion go to? do you know? were they planning on building a bespoke system or something?

 

[jimjamuk]

imagine going to a kick off meeting knowing there £1M worth of kit in stores ready to go for the customer (gov project) to decide that they ordered the wrong brand of kit and they were going with someone else. Sell that kit to the company for £1 (as the storage costs were going to be added for taking up the room) and then basically start again. Oh and then pop into the project a few months in that they expect that kit to be able to perform some cruicial function (which it cant). Redesign then order new kit and start again. Rinse and repeat

 

[Skeeter]

Can't help but think of iSOFT as I read this.

That's a blast from the past!

 

[Malt_Vinegar]

It would be £100k up front and then a monthly payment for the internet connection. Set up a central server, collocate at a datacenter for example with 1gbit link. Then you use existing web based medical records software instead of developing bespoke. The software would be at an additional cost as i have no idea how much they cost i estimate about £30-50k but maybe more. Set up a firewall vpn that can take that many connections. The additional cost would be digitising the old medical records and importing them in to the system, that is not included in the £100k.

This is gold, I can't believe you are building this concept system all for free!!

It sounds perfect, we should do it NOW!

 

[memyselfandi]

This is gold, I can't believe you are building this concept system all for free!!

It sounds perfect, we should do it NOW!

I like the way that he is serving medical records for 32,000 users (using Dj_Jester's figure) off a single non-resilient server.

No clustering (probably both local and geo- given the subject), no DR, no taking into account the resilient storage IOPs requirements, no backup infrastructure, no hardware and software maintenance costs.

Such a system would need to be available 24x7 so it can be queried in the case of a medical emergency in order to retrieve a patients records and hence must be very resilient.

And as for software licensing ... which can be many, many times your entire £100k budget.

Yes, £12Billion seems to be on the high side but we're not talking a simple system which you can do for £100k.

(Disclaimer one of my jobs is checking solution designs for technical correctness and supportability)