Thursday, June 09, 2011

HP Source Leaves Poor Taste

Last week the MD of HP in the UK, Nick Wilson (previously of IBM and CSC) made punchlines in the UK by claiming:

“They’ve canned [gCloud],” Wilson said.“They thought cloud was a bit too much nirvana, so in the short term, the projects that are being looked at are data centre consolidation.”

The outside world largely ignored the article - perhaps they thought gCloud was already dead and long since buried; the internal world suddenly asked questions like "we canned it? what about all the work we've done?"

Our HP MD went on to say

“If you want big savings… if you want to get into the billions, you have to go to the big companies with big balance sheets and big ideas,”

Ah yes ... the big companies with the big ideas. And the BILLIONS . They can deliver the other kind of nirvana, the real one. Oh. Wait.

There is, though, some confusion about what gCloud is or might be. Even government's own don't always seem to know :

Meanwhile, the G-Cloud strategy has been dismissed as unnecessary by David Wilde, CIO for Westminster City Council. When asked back in March [by a Parliamentary committee] whether the government’s concept of a nationalised cloud was a mistake for most government data, he responded: “Why have a nationalised one when there are so many privatised ones out there already?"

Returning to HP, I have it from an impeccable source - that means one that may know the truth rather than one who doesn't, as apparently found by HP - that they did indeed propose that they could save government billions - well £1bn to be precise - through the BIG idea of data centre consolidation.

After some rudimentary due diligence (of the sort, 0.1+0.02=?), HP have had to back away from that and offer, instead, some £120m of saves at best, and then only in return for extensions to their existing contracts. Those will not be forthcoming. As someone might have once said, a billion here and a billion there and pretty soon you're talking real money. Someone has banked that money already.

HP's response to the story was somewhat predictable, roughly paraphrasing "it's all a mistake, it's out of context" and so on, or, specifically:

HP have responded to the gCloud story by claiming that "Nick Wilson [actually] said the first priority for government was to look at datacentre consolidation, after which topics like G-Cloud would be addressed. As with many commercial clients, the process of datacentre consolidation is often a first step towards use of cloud."

Everything I hear today is that gCloud is alive and well. It is, though, a programme, not a thing. There isn't going to be a big cloud (in the sky) owned by government into which each and every bit of IT will be shovelled, dribbled or piled.

People inside government continue to work on gCloud and, whilst it's not without [some pretty significant] challenges, it's making progress. Ten years ago when I was at the centre of government, I would have done such a project with some pretty substantial seed funding from HM Treasury and I would have made a strong case for some kind of mandation - I'd have wanted government to get behind whatever the offer was and direct people to use it. That didn't happen then - cf gateway, DotP etc - and it isn't going to happen now. The difference now is that departments phone up the gCloud team every day looking for opportunities to join up - to save money, reduce risk, speed delivery and get something done. The pressure is on and departments are looking for ways to reduce that pressure.

My understanding is that gCloud is, amongst other things:

- A range of commercial offers, often in the public cloud (the privatised cloud as David Wilde quipped), though private cloud will feature too, covering everything from IaaS to PaaS to SaaS

- Services available initially at IL0 and IL2 (taking those at the very simplest level, that means, say, gMail might be IL0 and Office 365 might be IL2 - it's way more complicated than that inevitably but one of the biggest differences between IL0 and IL2 is whether the data is in the EU or not); IL3 will follow though that is substantially more likely to be in various private clouds (though not in a single "gCloud"). The latter examples aren't always true but they're true enough to be taken as precedence perhaps.

- Aimed at stimulating the widest possible market by lowering the barriers to entry for provision of services (decluttering the commercials as well as taking services as they are rather than with overwhelming government customisation) and so helping smaller businesses gain entry to the government market

- Architecture neutral. What's wanted is bare tin at IaaS, a range of suppliers putting capability on top of bare tin at PaaS and true services that are platform agnostic at SaaS. Government is buying services. They want those to be assured services - secure, reliable and performing to service levels (I hate the word performant, if it even is a word) and so on.

- Already making progress with email suppliers being evaluated at a pilot level and several more projects in the wings

I disagree with HP that rationalising data centres is the "first step towards use of the cloud" - gCloud is not about taking what you have and cramming it into an ever smaller space through virtualisation (though I do agree that is a useful, but separate step). Instead it's about moving services out of the behemoth contracts that government operates today and replacing them with services provided by others. In less than 5 years many of the big government IT contracts will come up for renewal and nearly all of those are over £1bn a year, much less under £100m a year and so in order to make a dent in those, something more radical than a bit (or a lot) of server consolidation is needed. Something far more radical.

This new model brings all kinds of problems including federated authentication, managing a multi-sourced service offer, bandwidth upgrades for internet connections in departments, the need to upgrade from IE6, lighter touch procurement, revised terms and conditions that support SMEs, a move from typically funding IT with capex to funding it with opex, TUPE (a topic I've raised before) and many more.

I will post about a few of these in the future.

Meanwhile, the cloud is very much alive and the gCloud programme continues. As I said above, not without its challenges and none of those are easy to solve tomorrow but the energy behind solving them is large and constant. What's important is to keep that focus when it would be easier to switch to other topics - especially as, inevitably, people come and go and attention wanders.

It would be a shame to let this one go given how often the goals of shared services, lower cost, more SMEs etc have been articulated. It needs to succeed. And the supplier community should get behind it, as should the wider government community.

Tuesday, June 07, 2011

Napsterising Government Data

I spoke with an organisation the other day who hold 2 petabytes of digital image information. They see that moving to 20 petabytes over the next 5+ years.

When you're talking that volume of data, in any current delivery model, you're likely to be locked in for the long term - you'll never move out of the data centre you're in and will be forced to add ever greater bandwidth and infrastructure capability to deliver the information. You will, of course, need to employ an aggressive strategy for dealing with the large amount of data that will be accessed frequently shortly after creation, occasionally some weeks after creation and then almost certainly never again.

I propose that the data be split up and stored (in multiple copies) at every possible [government] network node using a BitTorrent/Napster-like algorithm that breaks data into tiny chunks storing it in several of those network nodes and then reassembling it from those sources when its needed. It's possible that the storage equipment in each node will need to have the algorithm directly coded within it.

Resilience is inherent and performance is improved because of the spread. Sure, you need more storage than you originally had (but you needed 2x your storage to provide for DR in the first place) but you need a lower grade of storage to achieve the same performance and throughput.

This should result in overall lower cost for storing and managing the data and a greater performance for those recalling the data. In this case, by the by, the data is X-Ray images.

Monday, June 06, 2011

It's Been A While

Ages, in fact, since I posted a flip chart image. This one, well, I don't think you could make it up:


Thursday, June 02, 2011

One Blue Badge, One Bl-ue Badge ...

At a meeting today we were kicking around thoughts about services that should be online and easy to use and that probably weren't or, at least, aren't consistently available. We quickly alighted on the disabled persons blue badge. Instinct said that if anything was going to have a consistent national process, this would be one (after all, many e-government ships have foundered on the rocks that are housing benefit processes). I was grateful that direct.gov let me type a post code and then routed me to the right place to see how to apply. So:

My local council, Hammersmith and Fulham says you qualify if you:

  • receive mobility allowance or the higher rate mobility component of disabled living allowance
  • receive a war pensioner’s mobility supplement
  • are registered blind
  • have a permanent and substantial disability which means that you are unable to walk or have very considerable difficulty in walking
  • have a severe disability in both upper limbs and are unable to turn by hand the steering wheel of a vehicle even if that wheel is fitted with a turning knob
  • children under 2 who use bulky medical equipment
There is, though, no online process and no ability to download a form so I can't see how they test these criteria.

Somerset has slightly different criteria - very similar but more detailed and these distinguish between automatic and following assessment:

You can get one automatically if you:

are registered as blind under the National Assistance Act 1948

you receive mobility allowance, or the higher rate of the mobility component of the Disability Living Allowance

you receive the War Pensioner's mobility supplement

you are a severely disabled service person or a veteran who is compensated under the new Armed Forces Compensation Scheme (AFCS)

You can get a badge subject to further assessment if:

you have a permanent disability that means you can't walk or find walking very difficul

tyou have a severe disability in both your upper limbs and you regularly drive a motor vehicle but cannot turn the steering wheel by hand, even if the wheel is fitted with a turning knob

you are a child under the age of 3 who needs to be accompanied by medical equipment which cannot easily be transported, or who is affected by unstable medical conditions and may require emergency treatment.

Their process can be completed fully online - I didn't have details of a local address so couldn't try it out but it looked like it would work, if that means anything.

I then went to Buckinghamshire's site (I'm running out of post codes that I know) and was impressed. They have an eligibility test online. I failed the test the first time because I didn't confirm that I actually lived in Buckinghamshire. There follows an online form that looks very much like that in Somerset.

Finally I arrived at the Lambeth site and, at last, found the 20 page form I would need to fill in were I actually eligible. A 20 page form. For a Blue Badge. It's not even the large print version.

The interesting bit is how the checks on eligibility are carried out. Lambeth requires proof of all sorts of things - address (the usual utility bill etc), pension etc - as well as photographs, a cheque for £2 (soon to go up to £10 I gather - another hit to inflation).

It strikes me that with some up front authentication, it wouldn't be hard to test all of the other criteria from a single central capability (let's start with England, I get the sense that the rest of the UK will be more challenging). Yet here we are, 6 years after "2005, 100% online" with a service that could be used by up to 2 million people - perhaps the very ones who would most benefit from a simple way of getting hold of something that would be genuinely valuable to them ... and it's not there.

I then found this statement, from February 2011:

We are establishing, with local authorities, a common service delivery project which could deliver operational efficiency savings of up to £20m per year, help to reduce and prevent abuse and improve customer services. The project will also make available an on-line application facility which should result in faster, more automatic renewals for people whose circumstances do not change between renewal periods.

I haven't heard that anyone is actually doing this. £20m of saves sounds worth having, assuming those saves can be delivered for less than £40m! Anyone know if this is underway?

Wednesday, June 01, 2011

How is Nokia Doing - Update

On April 16th I wrote

"I'd like to see Nokia unveil a clearer plan - perhaps like my prescription or an entirely different one. Waiting until 2012 before we see Windows phones doesn't seem like a good idea - and in the interim the only headlines will relate to lay-offs, budget cuts, sales of units and other negatives, all of which will drive away customers and so put the stock in negative territory. Mr Elop has a big job ahead of him. Let's hope he's not like the CEO of Microfocus who departed after less than a year on the job this week, with the stock pricehaving fallen 41%. Instead he'll be hoping he's more like Bart Brecht who say the shares of Reckitt Benckheiser lose $2bn in value upon the announcement of his departure."

Yesterday the stock fell to a low of $6.79 before recovering, a little, to $7.02 (it was at $8.66 when I wrote the paragraph above). The headlines continue to be negative and the company isn't putting out a good story - instead they know it's going to get worse, withdrawing guidance for the rest of 2011 and only citing "increased confidence that the Windows Phone will ship in Q4 2011".

Next week is WWDC with iOS5 at the very least (as well as iCloud) if not a new (or upgraded) iPhone; doubtless the Android team will put the boot into Nokia not long after that. It will, certainly get worse before it gets better. And maybe it won't get better. That would be a shame.