Tuesday, 11 April 2017

JD Edwards 9.2 to live on… and on… and on…

image

Collaborate inspired content

The announcement, incase you missed it, is that we are going to have JD Edwards 9.2 until 2028.  Wow, the propaganda that went with the delivery was pretty awe-inspiring, but honestly what does this really mean?

Is oracle trying to squeeze out the last bit of life from JD Edwards for the smallest investment possible?  We were very used to the traditional major release model, that is https://shannonscncjdeblog.blogspot.com.au/2016/10/jd-edwards-release-datesmakes-me-feel.html every 3 or so years we were getting a major applications release.  This was good, it kept clients on their toes, kept everyone informed of the latest updates.  It was a self promotion and marketing exercise at least, as all partners went out to their clients and told them why they needed to be on the latest release….

image

I must be honest, I’m in two minds about this announcement.  I can see a lot of positive things about it – but also negative.  I’m a half glass full kind of person, so lets start with that:

The full half of the glass:

If we can change our mindset to be one of constant change and constant innovation, this announcement is cool.  We must however make the change and bring our clients along for the journey.

I always think about this as if I’m providing JD Edwards as a service for my clients (as I do) and how I can provide them a constant platform – which includes innovation.  This IS a difficult thing.  AWS gives me some amazing ability to do green / blue deployments and to test code changes slowly and constantly – but I need to build this for JD Edwards.  JD Edwards does not do this well natively.  For instance, if I was hosting JD Edwards for a client, I could easily deploy (with them knowing) and new pigeon pair web server and app server on the latest tools release alongside the current tools release (green / blue).  I’d carefully monitor performance, issues, logs and more and eventually phase all of my users over to the new tools release and they probably would not know any difference.  This ability to do constant change is made easier with the constraint less compute environment that I’m using to host JD Edwards.

The above is a small example with a big change (tools release), but changes get MUCH bigger – think application release.  Now this hypothetically could be done in a very similar way…  But, data, my single source of truth is going to be my challenge.  I could easily write triggers and routines that would synchronise (or run TC’s inline) to keep JD Edwards running between the two releases.  Wow, imagine that.  Go live for a group of users / locations…  Test and continually improve and deploy – possible.

Read all of the above and you can see that this is actually WHAT WE EXPECT.  We now expect our programs to update automatically, we expect the latest support of platforms and browsers and more importantly mobile device operating systems!  Cloud or more specifically SaaS has completely changed our expectations of large software.  We do not want to do big upgrades and create big disruptions, we want to do small and consistent upgrades with no disruptions.  I think we can do it.

If you are internal IT or you are a managed service provider, think of your JD Edwards instance as SaaS and think how you can give your customers a consistent and contiguous environment that is always up to date…  You can!  If you make the paradigm shift mentally, you can start to think creatively about how you are going to do this.  Oracle (in making this continuous improvement announcement) have forced us to think about our ERP differently, and for the better.

You are getting an environment (well this is what it feels like), where you can more easily provide a managed service.  Wow, is that what oracle is going to do eventually – hmmm, I’d think so.

There are some repetitive tasks that you will need to get better at to ensure that you are ready for continuous innovation:

  • get better at retrofit
    • code better
    • modules, reuse and more
    • know the cost of your modifications and bring it forward if there is real benefit to the business
  • ESU’s all the time – put them on a schedule
  • Get your underlying technology ready to support continuous change
    • blue / green deployment
  • invest in automated regression testing
  • monitoring is critical (how about ERP Analytics)
  • Performance testing is important

At the end of the day, JD Edwards is going to play better as a SaaS product – though it will still need to be managed closely.

 

The empty half of the glass

We need to also think critically when we get an announcement like this too.  I had a feeling that the software was being put out to pasture when I heard that there was going to be no more major releases.  It felt like this was not a change going forward, but a change to stagnate, but that is because we all resist change – it’s natural.

We’ve seen how it’s really not too easy to sometimes run blue | green deployments with some of the releases that have come out lately.  Trying to use 9.2 tools on 9.0 apps is terrible, we really hope that things are going to be architected and released in a way that can be consumed continuously.

This is a shorter paragraph, as I think that the message is overall positive.

What they hope to deliver continuously:

These shots were taken at the partner session at collaborate, so they are completely immersed in a number of safe harbor statements.  

imageimage

 

Wow, so that is a lot of enhancements that are going to be delivered into 9.2 as continuous items to 9.2.  All of the above is going to be bolted onto the current 9.2 applications release.  We are going to get tools releases, but applications release will stay at 9.2.

Conclusion

If JD Edwards can continue to innovate strongly, but continuously – all we need to do is ensure that we are also ready to consume this at a rapid pace and allow our users to benefit from it. 

The JD Edwards cadence of innovation has been exceptional, they are continuing to provide the tools to enable a business to make their digital transformation to a maturity level that is appropriate for the reason that they exist.  This is a deep statement, but the level of digitalization that an organisation can achieve is governed by what the company does and the reason that they exist. 

I look forward to continuing to architect systems for my clients that are future proof, that will embrace this announcement and allow the customers to actually benefit from it now and onwards to 2028…

No comments:

Extending JDE to generative AI