The World According to Patterns and Non-Patterns

Share |

I spent the last two weeks on vacation, cruising down the Danube river with my wife Jeannie.  It seemed like, as with a lot of vacations, we fell into a pattern during our visits to a number of cities. Seeing the town center, palaces and castles, and eating.  And there were some things that didn’t fit the pattern, such as when we decided to have a good ol’ American hamburger in Budapest.

We all live with patterns and non-patterns.  The components of TM Forum Frameworx contain both.  Patterns are good, because, among other benefits, they represent a consistent approach to modeling.  Patterns are used to develop all the components of Frameworx, such as developing the Information Framework (SID), decomposing processes in the Business Process Framework (eTOM), specifying and developing TM Forum interfaces, and identifying and defining Application Framework applications.  But, there are always situations that don’t fit a pattern, which is also good, or we would be bored to death!  I call these non-patterns.

Patterns aren’t new or unique to Frameworx.  Many, if not all, of the patterns that we use have been around quite a long time and even pre-date my old bones.  Often members ask me why we picked the patterns that are used.  I won’t get into going through all of them but will provide a few examples.

From a SID perspective, the real world provided us with some guidance.  When we go shopping on the web, there is often a link to the specification that describes the object of interest to us.  This is one part of the EntitySpecification/Entity pattern – the EntitySpecification.  And, when purchased, we receive our own personal instance – the Entity.  Often the object of interest is a bundle.  There is another pattern that can be used to model this requirement, the Composite/Atomic pattern.  The bundle is the composite, which may be made up of other composites or bundles and/or atomics, the lowest level objects of interest  This also demonstrates that patterns can often be used together to satisfy a set of business requirements.

From an eTOM perspective, the category of object, or entity, often can be used to identify processes.  Forms of interactions, such as orders, problems, and such, move through a common life cycle.  They are opened or issued, reported upon, tracked and managed, and closed.  A more generalized life cycle for any object can be thought of as plan, acquire, use, dispose.  This is not to say that this is the complete life cycle.  For example, problems must be isolated, which doesn’t make sense for an order!  So patterns can be viewed as a starting point for modeling.

Which leads to also thinking about non-patterns where a requirement cannot be satisfied by a particular pattern.  I’ll provide just one example from the SID to make this point.  Currently the group of entities, called an Aggregate Business Entity (ABE), used to model a Customer, such as Customer, Customer Account, Customer Credit Profile, Customer Contact, does not employ any patterns.  Why – because the patterns that are used to model the SID do not satisfy any customer information requirements.

And, certainly, other patterns can be used by members when extending any Frameworx components.  I just would try to limit the number used or the resulting model will be much harder to understand.  Another point is that to satisfy a set of requirements, a combination of pattern(s) and non-patterns may be employed.

Off to Team Action Week Baltimore next week, and I’m certain another blog topic will be based on the week.  Talk to you again soon!


Posted 07-15-2011 9:13 PM by John Reilly

Comments

V.R. Sitaraman wrote re: The World According to Patterns and Non-Patterns
on 07-16-2011 2:00 PM

Interesting.  Francis was talking the other day about the need for history (temporal characteristics) - A time domain consideration.

Patterns make me think of something that is repeatedly consistent forgive the oxymoron. - A Freqency domain characteristic.

Combine the two we have the 1s and 0s out of which  the telecommuncation world. is built.

One cant help but think of duality. I have heard it occurs in many fields mathematics, control theory, physics, chemistry and so on...Duality between time and freqency domain. An impulse in freq is flat in time and vice versa. What is pattern in one is non-pattern in another. What you cant miss in one is easy to find in the other.

Can we in OSS exploit this.. One thought that comes to mind is fault RCA and SIA  arent these fundamentally breaks in patterns? Would we be able to view it in both domains, will be be able to reuse parts of tremendous wealth of information on subjects stemming from these into our field?

Can we then with statistical data standardise patterns and non-patterns for conformance and violation checks for fault?

christinawhite white wrote re: The World According to Patterns and Non-Patterns
on 07-18-2011 12:51 AM

<a href="www.cdrfence.com/">CDRFENCE | Fraud Control For VoIP Networks </a>

christinawhite white wrote re: The World According to Patterns and Non-Patterns
on 07-18-2011 12:52 AM

nice comments... good website.....

V.R. Sitaraman wrote re: The World According to Patterns and Non-Patterns
on 07-18-2011 8:29 AM

After reading Johns blog i am taken in by the relation of pattern and non patterns to life, and its application in our field and am realizing how much depth there could be in it.  

My Materials background has gotten a fair amount rusty,  but my roots connect patterns to diffraction patterns. Carrying forward Johns observations and Chirstinas pointer to its use in CDR anamolous call patterns, couldnt we carry it further for wider applicability?

Diffraction patterns are a result of constructive and destructive interference.  In addition to fault, isnt this what Traffic Engineering is all about too?  Isnt equal distribution (fair allocation ) translate to at least in part a semi-destructive interference (spatial, temporal) in your traffic frequency?

Wouldnt existence of diffraction patterns based on frequency immediately identify trouble spots where traffic is constructively interfere?

Device OEMs who have captive FABs would have amongst them experts in this field who knows the ins and out, a tea invitation and a request to look at our directions could help them help us throw a new light on traffic management, congestion control.

Then the interfaces  can take over and allow specification (just like in QoS) desired patterns (at various levels of granularity and hierarchy.

 In Materials once you have the sample thinned and reday to go, we have little say in its structure, and we use Diffraction as a characterization mechanism.

 SD1-44  diagram reminds me of Materials Crystal Structures.  Networks have nodes (atoms) and links(bonds) and exhibit structure. Structure-Property Relationship s are then derived. Here, in our field, the OEMs allow us to define our own structure, we go the other way alter the structure to get a pattern (or the absence thereof.)

  QoS has always to the limited extent i have been exposed to have been dealt with as queues, buffers and drop, scheduling rules. Simple  one dimensional structure. Why not explore (MFDFr is a crystal structure in itself albeit a very simple one) a three dimensional structure and related theories i am sure exists out there?

   Security analysus for suspicious patterns, the list seems to grow as you keep thinking about it.  

 Maybe NSF whose good graces have made the apparently impossible possible in many an area could look our way and fund such  unificationary interdisciplinary study.  Who else is better positions to run such research, than bodies such as TMF, a melting pot of OEMS in hardware, software, management, operations, billing all putting their minds together.   Telecommunication Networks are one of the most important Materialistic (pun intended) the world has got today and advances in it would more than compensate for any investments made on it.

Karen Keri wrote re: The World According to Patterns and Non-Patterns
on 07-18-2011 1:44 PM

Very Interesting observation, as this can go overlooked to many and just stand out to a few.  

V.R. Sitaraman wrote re: The World According to Patterns and Non-Patterns
on 07-20-2011 2:16 PM

And then the network with buffering at the ports for packets and server-client layer could be viewed as nodes and links with different "elasiticities" isnt it? That opens up knoweldge available in many body theory, statistical physics to our field, why, we can get very fancy and see if even at this macro scale there are quantum regimes, say if you can clearly pin point the QoS the futuer demand matrix feasbility to the planning  departments chagrin is uncertain and if you can clearly plan for future demand matrix, to the customers chagrin, the QoS in uncertain! In business terminology a volume-margin duality.

V.R. Sitaraman wrote re: The World According to Patterns and Non-Patterns
on 07-20-2011 2:35 PM

I was reading that one aspect that TMF is looking at is OSS/BSS convergence, never for once imagined that  BSS/OSS convergence could have anything to do with particle physics and QM, oops, i must be having Schrodinger turning in his grave....

But a bit more  realistically,  if mathematicians and physicts could grace wall street, why not have them look at OSS/BSS?

We welcome your feedback! To comment on this blog post please either Log-In or Register to the TM Forum Community

Paid Advertisement
About TM Forum
Introduction, History, Board, Management Team...
Membership
How to Join, Benefits, Member List...
Community
Community Home, Groups & Teams, Blogs...
Conferences
Event Calendar, Management World, Supported Events...
Training & Webcasts
Upcoming Training Courses, Upcoming Webinars, Podcasts, On-Demand Webcasts...
Initiatives
Cable, Enabling Cloud Services, Government and Defense...
Best Practices & Standards
Frameworx, Business Process Framework (eTOM), Information Framework (SID)...
Resources
Document Library, Case Studies, White Papers
Research & Publications
Business Benchmarking, Newsletters, Insights Research...
Copyright © 1988-2012, TeleManagement Forum. All Rights Reserved
Contact Us
Careers with TM Forum
News Room
Privacy Policy
Terms of Use
Sitemap