Welcome back!

I’m very happy and excited to be joining WSO2. The company has great leadership, great technology, an enthusiastic collaborative culture, and tremendous potential.

After working in the financial industry for about 10 years, I’m also welcoming myself back to blogging. I first started in 2004, when I was CTO at IONA Technologies, and more or less stopped when I took the role of Chief Architect, Investment Banking at Credit Suisse. Blogging was not really part of the banking culture, at least not in 2009.

This month I started my new role as CTO at WSO2, and am going to get back to blogging again. I see from the stats I should have started a few days ago, before the public announcement of my new role. The site should have been updated in advance. Well, call it a slow restart I guess….

I will be back soon with some thoughts about the industry, some observations about where we seem to be going, and of course how I see WSO2 in that context.

4 responses to “Welcome back!

  1. Not sure if this is the right place for this…
    Your thoughts on software, e.g.,
    – where is it going?
    – do we write too much? Should there not be more tools to generate or perhaps better tools to integrate what is already written?
    – how do we get software to use entire machines effectively, e.g., 64 cores, or more? What languages are best suited for this? Or do we need more?
    – managed code, e.g., Java, is this the future? with AI making big strides, will it not be possible to optimize runtime even more than today and also make use of available hardware resources?
    – microservices seems to be a bit of a misnomer as they appear to cover just about everything, from small to quite large?
    – it is said that a microservice must be scalable, stable, able to be replaced without shutting anything down, perhaps transactional, and many more attributes. It would be great to hear your opinion on this, including the integration of ‘legacy’ applications such as we worked on many years ago

    Enough for now and I hope this is the correct place for comments such as this? If not, please point in the right direction!

    Cheers, John
    P.S. It is wonderful to see you back ‘online’, Eric! I may be retired, but do still try to keep up with things technical

  2. John – sorry for the delay replying. And thanks for the kind words.

    Right about the time you posted this I was getting ready to travel to Aruba for a week.

    I hardly know where to begin to respond. The key thing about microservices – they are a specialization of SOA for commodity data center infrastructures. These infrastructures consist of hundreds and thousands of PCs networked together. These infrastructures are designed to tolerate the failure of any component at any time. Failure is planned for and assumed rather than treated as an exceptional event. To accomplish this the file system writes 4-8 copies of the same data to different PCs. If any one (or more) of them fail, there’s another copy available to the application. As you know this is a very different model to the way in which the VMS file system worked, which focused on storing a single copy of data reliably in a single place on a disk. Yes, some of the disks were replicated or dual-ported, but this is different than writing multiple copies of the same data to multiple disks. Commodity data center infrastructures were invented for Web applications and basically follow the HTTP (REST) model for shared state – no persistent shared state or connections. To get the most out of this model typically requires apps to be rewritten.

    At a very general level, I think we are seeing trends for multiple types of databases and languages as an aspect of the specialization of the industry. Meaning the data stores and languages are being developed for a special purpose, rather than for a general purpose, as they used to be.

    Not sure about managed code.

    Something interesting though is all the code being deployed “at the edge” of the network, including AI agents in cars, Alexa/Siri, manufacturing devices etc. The biggest example is probably Microsoft’s Xbox statistics engine, which updates networked gaming apps in near real time

  3. Thank you so much, Eric! I see from you other post that you had a grand time in Aruba – I envy you the sunshine and warmth even if Winter had not fully arrived here yet.
    You description of Microservices was more than useful and the best I have seen so far – thank you.
    I have noticed the multiple new languages and DB types; most fascinating and something I am trying to keep up with. Graph DBs seem to be suited to a lot of what is going on too, however, they do remind me of CODASYL and sets…
    By managed code I am thinking of platforms such as.NET, Erlang and Java whereby the code is optimized based on flow patterns. Yes, they can be wrong as month-end is different to daily processing, but surely are more accurate than anything ‘static’ such as compiled code? Unless, I suppose, that code is purpose-written and in small pieces suitable for Microservices in edge computing especially.
    The IoT appears to be driving a lot of this too, so maybe I am wrong in asking about managed code.

    Thank you for responding to my ramblings, Eric! It is fine to be back in touch like this.

    • John, thanks – it was indeed nice to get away, despite all the testing and precautions.

      I’m afraid I don’t know enough about managed code to comment – WSO2 is in the process of developing the Ballerina language, this will be something very intersting.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s