No, Time, thou shalt not boast that I do change:
Thy pyramids built up with newer might
To me are nothing novel, nothing strange;
They are but dressings of a former sight.
Our dates are brief, and therefore we admire
What thou dost foist upon us that is old;
And rather make them born to our desire
Than think that we before have heard them told.
Thy registers and thee I both defy,
Not wondering at the present nor the past,
For thy records and what we see doth lie,
Made more or less by thy continual haste.
This I do vow and this shall ever be;
I will be true despite thy scythe and thee.

Sonnet 123

thanks again to the sonnet a day.

(photo via cc attribution licence. ms4jah thanks.

Most of my working career encouraged me to chant the mantra, “we are integrated.”  I have spent the last year or so thinking about integration in a more objective light. Integration for me has undergone a de-reification process.

I was doing a presentation about HR technology recently, and I was talking about the impressive growth of niche vendors in various HCM areas, such as recruitment, learning, performance management and succession planning. One of the IT folks in the audience was getting a bit agitated, this was chaffing against his integration faith.

My new view is integration is like weather. The statement, we have weather today, is accurate but not particularly useful. This is similar to the statement our system is integrated.

What is integrated with what? What is the purpose of the integration? What value does the integration bring? What overhead does the integration create?

Once you know what is integrated, you can rationally assess the value of integration. Integration can be of significant value, it can help break silos, reduce data capture and improve data quality.

But integration can be an excuse not to move quickly, it can hinder innovation and create overhead, it can be a reason not to do something new.

Sometimes retyping stuff is okay. Tracking 50 top executives in a succession planning application doesn’t require real time integration with 20 separate global HR systems. In this case, typing is probably the correct integration approach. If you are doing performance appraisals for 300,000 employees then you had better have a plan on how to handle organization unit changes. This will involve a sophisticated integration framework with your system of record. Both of these are forms of integration.

When evaluating software view integration rationally, don’t put it on a pedestal, but don’t dismiss it. Understand clearly what is being integrated, and what the value of that integration is. Integration doesn’t trump functionality. Without the right functionality, integration is not worth much.

But also beware of those that say integration is easy to do.

(photo mine, use with cc attribution licence) Hotel Fusion, San Francisco.

I’m just back home from a trip to California. I attended Oracle Open World, and had several other meetings in the bay area. 

I saw the Keynote presentation, and on the next day  I spent an hour or so with the leaders of HCM product Fusion product. Colleagues of mine here at Gartner have spent more time getting a deeper look, so I will leave the detailed product analysis to them for now. No doubt I will learn more over the next few months, but at the event I was more focused on getting to grips with PeopleSoft 9.1 and catching up on progress of EBS 12.1.

I have read quite a bit about Fusion over the last few days on twitter and on blogs, some of it thoughtful, some of it wishful and some of it rude. 

My thoughts.

Larry Ellison spent most of the keynote talking about hardware, with the odd barb aimed at IBM.

(photo via Oracle’s cc flickrstream. thanks)

He left Fusion right until the end.

It was enough of a look to show that it is relatively close to being done, or at least significantly further along than halfway. Larry outlined the scope, and mentioned some of its innovations. The demo was adequate without being awesome. It set some expectations, but it didn’t promise a new form of sliced bread. What Larry didn’t do is pontificate on how much revenue this thing will do. No suggestions of 10,000 customers, or billions of dollars of revenue. He didn’t promise much at all, other than a vague some time next year. 

He generated interest, but not enough to put a freeze on PeopleSoft 9.1 and EBS 12.1 upgrades. 

It would have been nice to see a bit more, but Oracle is in no rush. Oracle has the Sun acquisition to complete and digest, so I’d expect Fusion get more of push once Oracle has Sun on board and aligned.  It wouldn’t surprise me if Fusion comes shipped on its own special piece of hardware.

Cactus

via Flickr, the cc licence of Rodolfo Cartas thanks.

In this architecture, everything is from one vendor, and integration with third party applications is rather difficult. Typical ERP /HRMS pitch of the mid-nineties. Why do you need other software? We can do everything.

Sunflower

via Flickr,  the cc licence of C.S. 2.0 Thanks

Big core system, running most of the processes, with a series of smaller, tactical solutions interfaced around the edges.  Typical HR IT architecture of many ERP-Centric organizations today.  ERP runs the core transactions, with bits of SaaS tacked on around on the edges.

Daisy

via Flickr,  the cc licence of law_keven Thanks

Small core system on premise, but most of the action takes place in the systems around the edges. Increasingly common as SaaS vendors continue to deliver richer functionality. Some challenges with integration, as there are many applications trying to connect to the core. 

Rose

via  Flickr, the cc licence of Gertrud K. Thanks

No significant core system, SaaS petals dominate.  Still very rare, but we expect to see more of these, challenging the traditional core and peripheral model. 

What sort of flower does your architecture represent?

(photo CC 2.o attribution, thanks to g-hat!)

World leaders are gathering in Pittsburgh to discuss banking reform and other pressing matters. According to the Guardian,  the discussions are likely to be rocky.

European leaders appeared to be on a collision course tonight with Barack Obama and Gordon Brown after Angela Merkel, the German chancellor, warned that the G20 summit must not be diverted from clamping down on bankers’ bonuses and hedge funds.

The article continues.

Sarkozy has suggested that bankers’ pay should be capped at a certain percentage of their institution’s assets or revenue.

Fredrick Reinfeldt, the Swedish prime minister and current president of the European council, promised a “specific discussion” on bonuses including proposals for individual caps on bankers’ bonuses, that bonuses would be linked to achievement and not given if there were losses, and that there would be transparency on precise decisions taken by boards. “We from the EU will ask to be very clear on that” he said.

Putting aside the ethical and political debate, if Fredrick and Nicolas have their way, this would particularly riveting for anyone in the business of HCM software.

It looks to me this is a demand for an integrated employee goals / performance management, compensation and incentive compensation system that also integrates into a corporate performance and risk management system, combined with a significant dose of compliance reporting. 

I came across the story of James May’s Lego house this evening. James May is one of the fellows on Top Gear, and he is having some success in branching out.

As the Inqusitr notes.

Top Gear presenter James May is giving away a free two-storey house. The only drawbacks is that the house it made out of Lego and you have to pick it up by Tuesday.

James May and 1,000 helpers has just built the world’s first full-size Lego house using 3,3 million Lego bricks. It even comes with a shower, working toilet, toaster, kitchen utensils and a bed.

The 20-ft tall house is built in Denbies Wine Estate in Dorking, Surrey, UK, but now the vineyard needs the land back to harvest its grapes. If the house isn’t removed by Tuesday at 8.00 AM, it will be hacked into bits with chainsaws.

According to James May, Legoland UK was supposed to take it to the theme park in Berkshire, but its too expensive to move. Legoland has critized James May, because he hasn’t consulted their model-makers on, how to built a moveable house.

Lego has donated the bricks to James May, so the house or bricks cannot be sold or used as a public attraction, but only given away for charity.

photo via the cc flickr of TchmilFan thanks!!

Now, totally ignoring my own advice above about Lego metaphors, this has a parallel in enterprise software, and that chestnut, SOA. SOA isn’t just about the building blocks, it is about how you actually put them together. Even flexible materials fail if they used incorrectly. Modular can become monolith, and then you need chainsaws…..

There are decades of science, engineering, standards, governance and accident that have created the protocols that make the Internet work. Its openness, flexibility and to use Zittrain’s term, generative nature, are in a  large part  because the Internet isn’t owned by one company.

When you shorten a URL, you are no longer relying on the "Internet" to direct readers to content, you now rely on the provider of the shortening service.  URL shortening today is a very useful feature, but it is not generative. You create a single point of failure, and you now rely on that company continuing that service, and that in turn depends on the capricious nature of the market. It makes navigation proprietary rather than open. It creates a taxation point where previously there wasn’t one.

The post from the CEO is very illuminating.

tr.im did well for what it was, but, alas, it was not enough. We simply cannot find a way to justify continuing to work on it, or pay its network costs, which are not inconsequential. tr.im pushes (as I write this) a lot of redirects and URL creations per day, and this required significant development investment and server expansion to accommodate.

tr.im has thousands and thousands of users, creating tens of thousands of URLs per day. But, we were a little surprised to learn, *no one* wanted to take it over. We quietly contacted a number of people within the Twitter development world, and nobody wanted it in exchange a token amount of money. No one perceived any value in it, or they wanted to operate a shortener under a differently branded domain name.

And, users will not pay for URL shortening, and why should they?

And, the data that tr.im generates — the hottest links that people are sharing right now — is all well and good, but everyone has this data. tr.im gets hit by countless bots every day farming this data to create and operate websites such as tweetmeme.com. So, *everyone* has this data, meaning it is basically worthless *by itself* to base a business on (as bit.ly and others are attempting to do) at least in our humble opinions.

And finally, Twitter has all but sapped us of any last energy to double-down and develop tr.im further. What is the point? With bit.ly the Twitter default, and with us having no inside connection to Twitter, tr.im will lose over the the long-run no matter how good it may or may not be at this moment, or in the future.

So, in summary, there is simply no point in continuing to operate or work on tr.im, and we are moving on to greener pastures. We appreciate all the support and kind words about tr.im we received over the past 12 months, but change is ultimately good, and bit.ly can more than accommodate your URL shortening needs.

 

My personal view is that URL shortening shouldn’t be a product, it should be a standard. HTML works precisely because no one company owns it. I’m not sure that URL shortening should be any different.  Today I view URL shortening as ephemeral. At best a scribble on a scrap of paper. If you need to bookmark something for keeps, keep the full URL. But I’m not sure that I want to see any company control an increasingly important part of how we navigate the Internet.

My regular readers will know that I’m doing lots of research into the impact of social software and user driven applications in an HR context, but there are some core applications and processes that need to be non-negotiable. Senior HR management need to put the discipline and governance in place to drive standard system usage. It doesn’t happen by magic.

It may well be that the earlier implementation of the standard HR system didn’t meet user needs, and the decision to re-implement makes sense. They have an experienced partner this time around, who has a pre-configured solution that has a good industry fit. The standard software’s latest release is improved.  They will use a proper formal project methodology. All goodness.

But I heard a couple of things in the discussion about the re-implementation that worried me in particular.

1. HR are too busy to dedicate resource to the project.

  If the user community for the application can’t dedicate resource, then don’t do the project.

2. The key user can only spend one day a week on the project.

One day a week on a project means you have a spectator at best.  Take the key people and put them full time on the project. Make their careers, bonuses and corporate happiness dependent on the success of the project. You need committed, not involved.

3. The key user is really technical. He can build the most amazing stuff in Access and Excel. It is great that you can find someone in HR that is interested in technology, but if he/she is the one who has built their empire in excel, you have some significant work to do to make them the champion of a standard application.

It sounds so obvious, but if this the system that will be supporting day to day HR activities for the next decade or two, then HR need to get their best and most motivated people on the project. Otherwise, IT should be spending their time elsewhere.

photo via wikipedia. thanks!

This weekend we did some camping, just across the border into France, and very close to part of Maginot Line. Next time I visit that area I’m going to take some time and actually visit a couple of Ouvrages, go inside and see the working railway and the logistics of the thing. It is fascinating engineering and architecture, made all the more impressive by its absolute failure to deliver on its promise.

I normally dislike using military similes and metaphors in business language. It belittles the horror of war. I’m not in the trenches, in the line of fire, or going over the top, luckily. But there was something about the Maginot line that made me think about the enterprise software industry.

I suppose one could use the Maginot metaphor to describe neo-client/server architectures, but I couldn’t help thinking about software maintenance. The assumptions about maintenance made 10 years ago don’t necessarily hold today. While leaders of large software companies crow about their maintenance margins, images of the Maginot line keep coming into my head. 

Take the case of  the newly named Archer’s Road. (via the Guardian)

A street in Sheffield that has been the butt of jokes for many years has finally won a battle to change its name to something less … behind the times.

Residents of Butt Hole Road long ago stopped seeing the funny side of the legions of titterers taking pictures of themselves with their pants down next to the road’s sign. After clubbing together to raise the £300 necessary to pay for a new sign, the local council has agreed to name the road Archers Way, in honour of its half-mile proximity to Conisbrough Castle.

There are many cases when a change in name makes good sense.

But why is it that some software companies keep changing their product names every few years? Do they understand and care about the pain, irritation and cost it inflicts on their customers?  Who exactly do they think they are helping other than the brochure department? Do they realise the vast forests of systems documentation that are made obsolete? The hours wasted doing find and replace in Powerpoints, and worse in the application code itself? The helpdesk and partner confusion?  The environmental impact alone of a large vendor changing product names is material. I wonder if I could plug that into a carbon footprint calculator? I bet it would be equivalent to a few jumbo jets or negate the impact of a newly minted LEED compliant building.

I’m beginning to think that we need to start referring to such applications as the application formerly known as…and…and before that….and before that….and originally. When road names change and country names change for the public good, it is normally because the people living there demand a change. I don’t know about you, but I’m not seeing lots of enterprise software customers clamouring for the renaming of the systems. Most of them still call them by the original names anyway.