Future directions of EPTS and an online magazine

[rainer93138]

Opher Etzion as the chairman of EPTS (Event Processing-Technical Society http://www.ep-ts.com/) has published a report on the all members conference call of EPTS http://epthinking.blogspot.de/2012/07/m ... azine.html we held earlier this year. One of the ideas discussed was to create online magazine as the major activity of EPTS. After some delay, in the last few weeks we have advanced in discussions among those who expressed interest. There has been a keen discussion (conference calls and Email exchange) with some supporters and some concerns.

Here is Opher's report on background, and reminder about the history and current state of EPTS:
http://epthinking.blogspot.co.il/2012/0 ... azine.html

It was also discussed that everybody interested in this discussion should contribute to this thread on the complexevents-forum.

[rainer93138]

Let's start the discussion which has already started between some of us - a little bit behind the scenes and not visible for all interested EP-TS members. Because we decided that everybody should publish his/her comments directly, this was my first comment, 7 July 2012:

IMO, this is the wrong algorithm (Ed.: to think about a EP-TS journal at this moment). First say, what is EPTS, who is "we", who is in the EPTS steering committee, who have elected them how, is this a democratic organisation, or is it a NoE (Network of Excellence) and a project with a budget, e.g. from the European Commission, with a responsible coordinator and a team with defined workpackages when what has to be delivered, or is it the hobby of some colleagues, etc.

BTW: EPTS is since 2001 the European Platform for Transport Science http://epts.eu/index.php?option=com_con ... 5&Itemid=6

If the actualization of the contents is not organized and guaranted, we should urgently switch off this EP-TS website www.ep-ts.com which was never used by the chairman (also this time Opher used his private blog to communicate this discussion) or the steering committee (as everybody can see it from "recent changes") and which only shows that there is no plan and no program and no traffic since a long time, fully outdated news, no SotA website ... Sorry, but we must be professional, otherwise it is a waste of time and effort.

Actually EPTS was never been founded yet really, we should do it now, perhaps with a different name and acronym?

(I will continue with my other comments later)

[rainer93138]

Roy Schulte has contributed to our discussion with the subject "Does anyone care about event processing?"
http://www.complexevents.com/2012/07/25 ... rocessing/

I fully agree with him, especially when Roy makes the allegory about circles and is saying:
"But business people and engineers don’t go to web sites, conferences or stores to read about circles or to buy circular products. They are interested in tires, wheels, pipes, gears and ball bearings because they see the use of them. Other than mathematicians, few people spend time thinking about the properties of roundness."

This is what I answered Opher regarding the concept of an CEP online magazine, 26 July 2012:
"Actually I do not believe in this concept and categories of rubrics (customer view, vendor view, practitioneer view etc.).
We would have to find rubrics according to subjects, application domains.
As I already told in the last mail discussion:

• People who are interested in intelligent or driverless cars, are not interested in HFT or AlgoTrading mostly. Or who is interested in Brain Computer Interfaces, is not ..., Or who is interested in Epigenetics or BioMed, is not interested in .... and so on
• So we would have to find editors and authors who are competent in such thematical fields and we have to tell the stories accordingly (for potential adopters, not only for CEP experts).

That's what I have tried to do since around three years with the different threads in the discussion forum of complexevents viewforum.php?f=13. We can see that there are very different numbers of views, although this community are mostly CEP people only. But actually we want to reach potential adopters, so such a journal must address special subjects/rubrics.

Of course we can also have a special CEP rubric or edBPM rubric, etc., where we discuss the special theoretical questions and concepts of CEP, like EPL's or challenges of context-modelling and so on..."

I guess this is what Roy means with:
"None of the scenarios listed above are about the kind of event processing that is the focus of the EPTS. Within the EPTS, “event processing” is understood to mean a kind of real-time analytics (or real-time business intelligence)."

And regarding Roy's expectation:
"I expect that the “buy” percentage of CEP logic will grow; that is, commercial general-purpose event-processing software will account for an increasing share of CEP-using applications and tools during the next five years. The event processing platform vendors will become smarter about how they package their wares. And user companies and software vendors building CEP-using applications will realize the benefits of buy versus build for a lot of their new projects. The rate at which this will occur is a discussion for another day, however."

In many cases when developers actually develop typical CEP applications, they are not familiar with CEP and do it individually hard-coded and invent all the stuff again like event detection, filtering, abstracting, correlation, programming event processing logic by an special EPL, etc. EPTS or a CEP online magazine must evangelize how to do things like driverless or intelligent cars (as an typical example fo an upcoming market) based on CEP platforms in connection with the Google car in Stanford (Sebastian Thrun's virtual university lecture viewtopic.php?f=13&t=322&sid=867f2c38fb5d9b18fa03e1793fb61cd1#p1438) or EPTS should contact automotive scene like Audi, VW, Daimler, Ford, ... which are already working on that. This is also true for the other domains mentioned above (BioMed, BCI, epigenetics...).

[rainer93138]

EPTS and why is a cooperation between industry and academia so difficult

A comment to Roy Schulte's last paragraph:

"The fate of commercial event-processing platform products is not really critical to the EPTS. The EPTS is a technical society that is primarily focused on the technology not on the commercial market. The EPTS could play a valuable role in educating mainstream IT people (architects, CIOs, project leaders, and business and system analysts) regardless of whether companies are building or buying their CEP logic. Or the EPTS could continue to function as a place for the exchange of information among a small group of event processing experts. Possibly it could do a bit of both. (As a biased observer, I’d also assert that there is a role for Gartner Inc., other analyst firms and web sites to examine best practices, application architecture and the relative merits of various vendors and products)."

http://www.complexevents.com/2012/07/25 ... rocessing/

Actually Roy has not included the role of academia, perhaps intentionally. Why should industry and academia cooperate? Although we have the discussion since years, typically in connection with the future of the DEBS conference series http://www.csw.inf.fu-berlin.de/debs2012/ as the allegedly main platform and social event of the EPTS.

Some problems to restart EPTS as it was discussed with some of you in spring this year - we agreed that we must discuss the points frankly:

1) Academia vs. industry or business

• Both groups work very differently for different goals and in different styles.
• E.g. to submit a paper to a industrial event/conference, is scientifically worthless.
• And vice versa: industry is not interested to read a scientific paper which is typically written in scientific style, in accordance with the typical outline (abstract, intro, state of the art, ... conclusions and future work, acknowledgments (your Mom, your prof, your colleague, some friends...), always the same and a bit boring, not imaginable in industry, both groups do "story telling" very differently...)
• Vice versa: academia is not interested to read an industrial paper which is often written in marketing style, unprovable assumptions or assertions, without state of the art discussion, undefined fancy marketing terms, etc.

2) Academic research vs. business applications

• Industry must earn money, must sell licencies, products, funded consulting, industrial and paid projects.
• Academia must - until today - generate fancy academic style CV's, fancy publication index, fancy PC memberships for our CV, invited highly (sometimes pseudo) theoretical or scientific style talks what does not interest industry, etc.
• Industry must contact or invite or generate potential adopters, customers, sell marketing budgets to them - what is a problem for a scientist to take marketing money and tell a talk at an industrial event or do such a paid research with the industry, etc.
• Academia must apply for "serious" funds from DFG, EC, ... other third party sources, which are ranked/rated and what is important for your evaluation; getting industrial money is more or less scientifically worthless or even a problem regarding your evaluation, fame and image - perhaps more or less depending on the country.
• The selection procedure regarding applying or submitting a paper or project proposal is more or less critical or even corrupted, actually fully outdated and much too slow (see http://www.sigmod.org/publications/sigm ... .korth.pdf and http://wp.sigmod.org/?p=488 what Tamer Öszu is saying) - also in many of the scientific events or calls for papers/proposals, PC-members are more or less worthless, doing nothing or are often not competent, are only listed, sometimes are even not asked to be a PC member, more or less a farce (see also viewtopic.php?f=13&t=289). This is often not worth the contribution or even the money of the industry as sponsers, no impact...

3) EPTS as a whole

• is mostly seen as not democratic, therefore it is sometimes forbidden by your enterprise to become a member in the name of your enterprise or university, etc.
• is mostly seen as an IBM organisation/business
• the process to become a member takes weeks, fully outdated process, beyond the state of the art of network platforms, based on paperware...
• no working working groups, no results, no goals, no visions what to reach when by which procedure or approach, etc.
• no cooperation with other national or international organisations like GI (Gesellschaft für Informatik), Complex Systems Society, EUCog, etc.
• no critical mass, stagnation since years, retiring colleagues, no young "replacements", no fresh ideas
• always the same business topics, dying financial markets and outdated applications like HFT, ...
• no future challenges and a consequent development of new topics in research and industry like intelligent cars, Human Enhancement technologies, robotics, brain computer interfaces, BioMedical, no systematical development of a research community for influencing national and international research programs, no systematic engagement in advisory boards of research programs, e.g. from the EC, US, Asia, China, Japan...

This is certainly not all and what is told open or behind the scenes by the colleagues since years, we discussed it massively already in Rome/DEBS08, ...
If we will restart, we must discuss such points frankly and very concretely with a strict agenda and plan how to make progress.
Some weeks ago I wrote Opher a comment to his last blog post about DEBS2012:
".... we often reinvent already finished discussions about CEP in the last years, even in one of the keynotes where the term "Complex Event Processing" was addressed again - what was definitely finished by an agreement in 2006 in Redwood City by the whole CEP community at the 2nd CEP Symposion (http://www.citt-online.com/downloads/2n ... opsium.pdf and http://www.citt-online.com/index.php?id ... 2&id4=more and http://www.citt-online.com/index.php?id ... r&id4=more, good times and a nice start of the CEP community): ...".
When EPTS invites keynote speakers who are not members of EPTS and cannot know the SotA of the community, we should at least discuss the planned contents of the keynote in advance and not risk to mislead a keynote in directions which are already sufficiently discussed by the community - except there would be a very good reason
http://epthinking.blogspot.de/2012/07/d ... rames.html

[rainer93138]

In the beginning of July (09 July 2012 17:01) I answered a mail of Roy Schulte (with 15 more on Cc of involved EPTS members or interested people).
Actually it was an open discussion and David Luckham suggested to discuss publicly on this discussion forum of the CEP community, but then not all wanted to be seen publicly and everybody should contribute by him/herself and should not be quoted if not agreed before (a bit strange for my understanding of a democratic process).

Roy has summarized his mails in his article http://www.complexevents.com/2012/07/25 ... rocessing/ and this is one of the sentences I answered:

...However, I have come to realize that far more IT professionals aren’t sure of what event processing is. Moreover, they don’t buy commercial event-processing platform products because they don’t know why they would want them. This is partly because of the peculiar way the terminology evolved, and partly because of the way that products that do event processing are marketed...

These are typically different questions for different interest-groups:

• Why is there not enough revenue? (actually this is not true for all, like TIBCO and SAP HANA, they allegedly had their best quarter or year ever, based on their CEP-products, why? and others not?)
• Is there still anything to investigate? or was all done 2002 with David's book "The power of events" where David has described and summarized the basic concepts - done by projects started in the nineties already?

The first question is interesting for business people, but not at all for scientists or universities. You must solve this with your marketing, sales and presales departments which have/had a lot of budgets for that. Typically there is no university involved in this mail-discussion, and obviously they would not be interested in such a EPTS online journal.

The second question is something what is not sufficiently and systematically discussed by the EPTS so far. It is interesting more for the scientists and depending on the time horizon it is mostly not interesting for the running and near-future business. Science would not get money from the business side to investigate such subjects because no fast enough ROI, so they must write project proposals for DARPA, European Commission, DFG and so on. Normally this is not interesting for the industry/business people, and even it's a risk for the science to include them in such proposals; the reviewers might judge why to support product development of Big Dogs like IBM, SAP, etc. by the money of the tax payers, invest your own money (e.g. Oracle or TIBCO would never join such project proposals). This happened a bit when we submitted the EASSy proposal (Event-based Adaptiveness of Service-based Systems) some years ago where we had the whole prominence of Big Dogs like IBM, SAP, Siemens, Software AG, Thales etc. on bord, proposal failed, also with the criterion "quality of consortium"; the consortium had together around 1 million employees... (BTW: other proposals with some of the same consortium members like iCORE started end of 2011, with a similar CEP based idea... just ask the reviewers why)

• Next point: use cases (always the same question and often discussed in the last years)

First we will not get a use case from the business side based on a real project, mostly strictly confidential, we always had to sign a NDA when we worked on such projects.

So, use cases can only come from universities, as prototypes or as abstract patterns or so, or as generalized PoC's or similar from your presales people from the industry.
This was actually our first EU project proposal idea 2006, DoReMoPat (Domain Specific Reference Models for Event Patterns, described in this paper http://www.citt-online.com/downloads/Re ... dix_v3.pdf. EU project was not started respectively withdrawn because of a missing coordinator in the last moment. It would be a nice project idea until today.

• Open research topics

2006 we started with the connection of CEP with BPM at the NY/Hawthorne, first CEP Symposium. Then we described this problem in a paper http://www.citt-online.com/downloads/In ... _Final.pdf and organized some international workshops about edBPM. This is done now, no research papers anymore, no open research questions as it seems.
But a lot of problems to introduce edBPM in real applications, like we have tried since 2010/2011, e.g. at Deutsche Telekom AG. DTAG also particiapted in some of our workshops. We documented these problems of real edBPM projects on complexevents-forum viewtopic.php?f=13&t=280, what the practical questions are actually. In the meantime, I am very skeptic that the reality of a large enterprise allows to realize the edBPM idea as a "concert of collaborating processes driven by the real-time analysis of events in the event cloud...". The reasons are not at all research-relevant, but are caused by the past or history and by the nature how such large enterprises do business.

Since 2009 we follow now the idea to transfer the edBPM concept to other non-business domains and to generalize it as Ubiquitous CEP (as we tried to focus in special threads on the complexevents-forum viewforum.php?f=13 like Intelligent cars, Robot Companions for citizens, Biology/Epigenetics, Brain Computer Interfaces, etc.) IMO, these are the real interesting future fields and we have already started to contact and cooperate with the specific communities like EUCog, VPH-FET, ECSS, ... This is a challenge what EPTS had to do very much more in order to evangelize CEP or ed(B)PM or U-CEP concepts.

But there would not be ROI within the next year, and so it's interesting for science but not so much for the business people.

Therefore, question again is how to handle these different interest-groups in a EPTS or whereever (NoE directed by a coordinator and a team and a budget e.g. from EU like EUCog, ECSS, VPH... (we would typically need at least 18 months to start such a submission) or democratic organisation with an elected board, budget from the members etc. like ??? international ACM, IEEE, or national GI are a different dimension, nobody would pay fees to EPTS)? and how to work systematically and with real results and progress? Deliverables?

[rainer93138]

This is interesting because of the BigData subject and what is DARPA doing:
http://www.datanami.com/datanami/2012-0 ... _data.html

"... real-time intelligence for soldiers on the field means gathering, meshing and analyzing data from multiple sources of diverse data types (from all those many new, complex sensors, text, image or video, etc) in addition to melding that data with the info from other systems, including for example, behavioral discovery and prediction algorithm-derived intelligence... A lack of integrated human-machine reasoning tools limits the ability of system to use operators’ knowledge and ability to understand complex data..."

It should be a nice challenge for EPTS to evangelize our concepts to DARPA and to the mentioned projects in the article.

Some of our uCepCortex group viewtopic.php?f=13&t=319 will attend the EC proposers day in Warsaw end of September 2012 http://ec.europa.eu/information_society ... fm?id=8636

Our proposal would not focus on military applications like DARPA of course, because this would not be a good idea in Europe, but we addressed e.g. emergency management and other domains.

A European flagship proposal in the EC's Horizon 2020 and Beyond is FuturICT http://www.futurict.eu/.

In October 2011 a similar EU project is started, iCORE, and Siemens or Software AG are members as the Complex Event Processing contributors, also as members of our other proposals we tried with EASSy, Cloudbox within the last years http://www.iot-icore.eu/

[rainer93138]

Another idea of an exciting working group of EPTS
viewtopic.php?f=13&t=252&p=1456#p1456

If we decide within the next weeks or months to continue with EPTS or with a similar organisation, I will come with a list of subjects and potential working groups sorted by the different fields of Human Enhancement Technologies, Brain Research, BioMed and Epigenetics, Robot Companies for citizens, Intelligent cars, etc. - based on what we have gathered the last 3 years viewforum.php?f=13.

[rainer93138]

In connection with the future of Event Processing Technical Society viewtopic.php?f=13&t=321 and why is a cooperation between academia and industry so complicated, this interview with Luke Muehlhauser is interesting http://www.exponentialtimes.net/videos/ ... uehlhauser. Luke Muehlhauser as the recent CEO of the Singularity Institute for Artificial Intelligence http://singularity.org/ refers to the concepts of Narrow Artificial Intelligence http://www.agiri.org/wiki/Narrow_AI and Artificial General Intelligence http://www.agiri.org/wiki/Artificial_Ge ... telligence.
He sees a lot of products within the next five years in the fields of Narrow AI. Five years is typically a time horizon which the industry is interested in - for instance what Shahrokh Mohseni from Deutsche Telekom told 2010 in Gent at our workshop "From Event Driven Business Process Management to Ubiquitous Complex Event Processing" (http://www.citt-online.com/CfP-edBPM-UCEP.htm).


See top left; "This was the tablecloth of WorldCafe table 6 after the brainstorming ... we first discussed why it's so challenging to bring the "Big Dogs" like Large Enterprises into the "FET 2020 and Beyond Initiative" - committed to the 1 Billion Euro budget - and how we could solve the problem. In the left corner above we see that an engagement and time horizon of more than five years is a serious problem for them. So we mainly discussed how to convince Big Dogs that there is a benefit of edBPM/U-CEP right now already and why they should join these technologies as first movers."
http://www.citt-online.com/index.php?id ... 0&id4=more

[rainer93138]

We already have started a special thread viewtopic.php?f=13&t=303 where we are discussing such a cooperation. In the meantime we already advanced and EPTS might establish a special working group in order to position and elaborate the role of CEP in this domain in accordance with the - since one year existing - VPH-FET roadmap and with the VPH Institute http://www.vph-institute.org/ which is chaired by Marco Viceconti.

A bit fun in the summer break:

Last year when we have written or at least contributed to the VPH-FET roadmap viewtopic.php?f=13&t=303&hilit=roadmap+vph#p1131, we raised the question "What the heck is epigenetics?"

In the meantime I had a discussion with the German medical scientist Dr. med. Eckhardt von Hirschhausen and he explained:
If a child looks like his or her father, then it's genetics.
If the child looks like the neighbor, then it's epigenetics.

We can use it as an necessary introductory term definition.

(Thanks to Opher Etzion for sharing this on Facebook!)

And w.r.t. the "reservations" of Marco Viceconti, another German scientist in the field of protein folding, Prof. Rainer Rudolph, made clear allegedly: Life is 100 % genetics and 100% epigenetics. So, no reason for conflicting positions.

[rainer93138]

Self-driving cars in 2019, report says.
viewtopic.php?f=13&t=252&p=1461#p1461

Such subjects would also be a very exciting transdisciplinary working group of EPTS where we could bring together all partners from industry and academia and from the different domains we would need - as suggested since 2010, to be updated a bit in the meantime http://www.citt-online.com/downloads/In ... stract.pdf