Sunday, November 13, 2011

Summary of Oracle Day Lithuania 2011


First of all, I would like to say a big thank you for all of you that came to my presentation. The audience hall wasn't big, maybe that was the reason why it was full. :)
Thus, following such a nice day, for those that were not able to come or chose alternative Oracle session, I would like to summarize the whole presentation in three schemes here.

1. The Hype Cycle
You have probably seen this picture before, nevertheless, I just want to point here out that we are still on the top of the hill of enthusiastic expectations about Cloud Computing. So, just please ignore all the hype about that, be patient until it stabilizes (same as SOA did).



2. The Service Scaling
Based on book “Cloud Computing and SOA Convergence in Your Enterprise, A Step-By- Step Guide" by David S. Linthicum, I have presented a recommended path for diving into the Cloud having a stable enterprise architecture on SOA background.
So, as you see in a picture below, David Linthicum first of all recommends identifying the services that could be deployed outside the enterprise firewalls and only after then deciding the exact plan for Cloud deployment.


For more information on this, you can read here.

3. The architecture level scaling
Going deeper into this subject and trying to find another option for Cloud, I have figured out that Cloud could be reached, as well, with an architecture layer scaling (my thoughts are presented more precisely in the picture below).


As you see, this is another option for safe (or with a small step-by-step) switch to the Cloud that could be accomplished through an architecture presentation layer scaling. This is not the best way for going to the Cloud but it is worth considering, especially when your architecture has a clear division among the architecture layers.

During the presentation I have also stated that Oracle is not yet ready for Public Cloud, analogical thought was stated a few days ago by Vesterli as well.

Thursday, October 20, 2011

SOA Cloud Architecture - Oracle Day Vilnius 2011


When I have received it from Oracle it sounded to me like “An old buzz, a new buzz, and something technical” and only after a few lonely evenings of contemplation have I managed to form some tough thoughts into one short presentation. It is hard to find the main thoughts from a very big perspective of a wide terminology but time is the winner here. If you don't mind looking again at this sophisticated title, you will see that it coined from three big words. Willing to analyze each of them you could spend at least half a life, e.g. “Democracy within China history”.

Nevertheless, I started word by word and finally I could say that you are more than welcome at “Vilnius - Oracle Day 2011” on 27th October from 2.25 to 3.05 p.m. .

Wednesday, October 19, 2011

Anonymity? Internet Users Need It


Yesterday I read a very good article where the author summarizes Chris Poole's opinion about anonymity and identity. Finally, I have found Chris Poole a strong and experienced IT personality, who expressed the whole Facebook and anonymity misapprehension. That is why I don’t have Facebook, Google+, Twitter, etc.

“We all have multiple identities. It’s part of being human. We’re all multi-faceted people.”

Monday, September 26, 2011

To Run or not to Run Without Measuring?


Before making a significant change it would be worthwhile to describe the final state of that change (How will you feel when you get there?) in case the change takes too long without giving a measurable result. Or, in other words, name your PKI's before you go live with your goals.

I had the same thoughts before starting regular jogging last summer. I thought that I wouldn't need to measure the obvious things, but what if the obvious is not the obvious?

Thus, the detailed question was: “How could I measure my jogging without interrupting the jogging?”

The answer came when I googled some modern jogging pages. Having read a few articles, I have chosen an all-inclusive, private “e-coach” package, with the following list of tools:
1. The mobile phone with GPS → the winner is “Samsung Galaxy i900”
2. The e-coach software → the winner is “Endomondo Pro
3. The hardware for heart rate measure → the winner is “Zephyr bluetooth



Pros:
1. Finally, I have found the motivation for buying a smart phone with Android :) Earlier on I was happy enough to use Nokia E51 (BTW, my new Samsung still spends in a drawer more time than in my pocket)
2. I was highly motivated for jogging during the first months
3. I measured my real physical condition
4. I have become very impressed about the e-sports market and the potential it has

Cons:
1. Now, jogging requires at least 5 minutes more to start :) The whole package of tools (including Samsung smartphone and Zypher) equals the whole package of questions: “Is it working?”

Resume:
I expect that one day the mobile technology will help us to live more pleasant life, and as we can see, this time is approaching, but for now, I prefer jogging then measuring along with jogging :)
I feel a real pleasure of jogging without extra tools.

Recommended solution:
Do jogging without extra tools each day and measure the progress only once per month.

Tuesday, June 14, 2011

Longer Means Longer

No matter what your religion is and what methodology you practice, this sentence is true for ever “longer means more time than normal” :)
I would like to dedicate this particular phrase to “Oracle SOA Suite” installation process on two different OS platforms. It sounds sophisticated, but who cares, dedication is still dedication.

I heard about a new SOA Suite 11.1.1.5 release of a version previously, and moreover, I had to check one issue with the Oracle Human Workflow engine (the SOA Suite component). So, the time for a new (renew) SOA installation has come :)

This time, as usual, I have started from making/copying the OS background. Here I don't have a full stop thinking process. The answer is rather standard: “VirtualBox+Ubuntu”.

There are many references “how to install particular SOA Suite on a particular platform”, so it is easy to find it on the Internet (BTW, it is a good practice to have the installation guide, or some white-paper called “step-by-step” that helps to avoid stupid mistakes).
Nevertheless, an interesting part of this story has begun at a place where I have never thought before it could, or more precisely, with installation of “Oracle XE DB Universe” (BTW, the recommended DB version for small deployment instances). It took me (as never before) an hour to make it stable. I will repeat that sentence: “the whole long hour”, and only after that I was able to move to the main SOA Suite installation steps (WebLogic, RCU, etc.) An interesting (thriller) story continues. After those SOA installation steps, my virtualbox has started to throw errors suddenly “Read-only file system”.

I have spent the next hour with this unstable virtualbox (it had the virtual disk error, etc.). Anyway, the result is that I have decided to give up and to switch to an old package of mine which is still worth trying to use when I get nervous: “Vmware Player and Windows 2003 Server”.

Oh yes, .... Windows :)

And then, based on Oracle recommended “Quick Start Guide for SOA Suite”, I took a very quick tour with SOA installation on Windows. It went without a single error, and it took me … not more that an hour.

Resume (2h>1h)
Longer means longer, no matter what your OS religion is :) When time is an issue, there is no time for religion :)

So, the very last recommendation:
When you have to install some proof of concept platform for just one test and you are not sure about the installation process, make use of Windows OS. It has more users so it should be tested widely. Shorter means shorter, popular solutions has more testers :)

And what about production environment?
Hmmm, the answer is simple: "it depends" or in other words: "make a new decision" ;)

Tuesday, June 7, 2011

Some Thoughts on IdM, SSO, SAML and CAS

We are having a hot summer time here, so sharing clever thoughts has become challenging for me. Nevertheless, I would like to write some words here, in short: "Identity Management".

Based on my last experience from diving into the world of unknown, or in other words: "the experience from hardening my knowledge base", I'm starting to redefine my learning principles. Earlier I had an opinion that the best, yet structured, way to learn something big and new is to find a new book (bible) and just to read it from cover to cover. And only after such a hard reading, with hands-on exercises, one could be able to form some conclusions and feel more comfortable in that new area.

I followed this path last time as well and I have read the whole book: "Identity Management: Concepts, Technologies, and Systems".



And ... I don't feel much more clever now :)

The fact is, the book offers an in-depth understanding of how to design, deploy and assess identity management solutions. It provides a comprehensive overview of current trends and future directions in identity management, including best practices, the standardization landscape, and the latest research finding.

But this book is to much theoretical and overly one-concept-oriented. Note that reading this book sooner or later you will punch the head against the wall called: "SAML" :)

I'm not saying that SAML is bad or even that it is not worth reading about, etc, I'm just saying that the book, the concept of IdM, has to narrow description (you could learn even more about SAML from wikipedia and its references).

I'm just asking myself silently, where are more chapters about SSO, IdM solutions, protocols, e.g.: CAS, JOSSO, Athens, OpenAM? ;)

After finishing this book I was still hungry for knowledge about the IdM (especially about the SSO solutions), so I went googling for more. I was more than happy about the results finally. I have found plenty of articles about CAS usage and what is more interesting, I have come across the "CAS and APEX integration" tutorial.

I have done the whole example and I must say that CAS integrates smoothly with APEX. Moreover the author gives more good advices for APEX, open web solutions, hardening, e.g. using ModSecurity.

Resume:
Never stop searching :) and nevertheless the bad experience from the last book, next time I will try to
search for better "bible" ;)

P.S. If you can do something with APEX smoothly (in this case CAS integration) you will be able to do it with Java and so on.

Disclaimer: I did not get paid to review this book, and I do not stand to gain anything if you buy the book. I have no relationship with the publisher or the author.

Thursday, May 19, 2011

Even the Content Has the Strategy

I read Janus Boye's blog regularly (BTW he is a very good writer)  and I have recently come across an interesting article "10 European content strategists to watch" which made me interested in "Content Strategy" and I have decided to read a recommended book there "Content Strategy for the Web".

After reading this book, I don't feel like an expert on "Content Strategy" but I will try to introduce this concept for you in short sentences.

Thus, the best way to start a clever explanation of the subject is to use Wikipedia-based sentences.
So, Wiki says: .... Content Strategy is "the practice of planning the content creation, delivery, and governance" and "a repeatable system that defines the entire editorial content development process for a website development project."

It would be clever enough to stop here, after a few words "... content creation, delivery and governance" ;) but if you are still interested in this topic and don't have time to study the whole book, you could only read a short article called "The Discipline of Content Strategy" written by the same author.



Still hungry for getting to know the Content Strategy?
OK, you could go and read the whole book, it is really worth reading. The language is simple and yet the author has this cosy humour. The only minor disadvantage which I have found there is missing chapter numbers (I like mapping logical structure and it is easier to do it with logical numbers).

During the journey into the knowledge world of content strategy, please note the main steps for the Content Strategy process:

  • Audit (What content do we have?; Is it useful?), 
  • Analysis (Objectives, assumptions, risks, success factors)
  • Strategy (
    • What content do we need to create? Why? 
    • How will the content be structured?
    • How will users find the content?
    • How will we get from here to launch?
    • What's next once the content is "out there"?).

Disclaimer: I did not get paid to review this book, and I do not stand to gain anything if you buy the book. I have no relationship with the publisher or the author.