Discipline means "Say what you do, do what you say, and prove it". Quality Management Guru Edwards Deming has implemented operationally through his PDCA's wheel (Plan Do Check Action). It is only recently that Quality Management Methods now enters in Software Engineering - they're only at the beginning so that very few firms have in fact enter the phase 2 described below. You can inspire from that. http://www.developeriq.com/Magazinestories/03sep30CMM.php3 The following describes the five maturity levels of the SA-CMM, highlighting the primary process improvements made at each level. 1. Initial - The acquisition process is characterized as ad hoc, and occasionally even chaotic. Few processes are defined and success depends on individual effort. For an organization to mature beyond the initial level, it must install basic management controls to instill self-discipline. 2. Repeatable - Basic acquisition project management processes are established to plan all aspects of the acquisition, manage requirements, track project team and supplier team performance, manage the project's cost and schedule baselines, evaluate the products, and successfully transition a product to its support organization. The project team is basically reacting to circumstances of the acquisition as they arise. The necessary process discipline is in place to repeat earlier successes on projects in similar domains. For an organization to mature beyond the level of self-discipline, it must use well-defined processes as a foundation for improvement. 3. Defined - The acquisition organization's acquisition process is documented and standardized. All projects use an approved, adapted version of the organization's standard acquisition process for acquiring their products. Project and contract management activities are proactive, attempting to anticipate and deal with acquisition circumstances before they arise. Risk management is integrated into all aspects of the project, and the organization provides the training required by personnel involved in the acquisition. For an organization to mature beyond the level of defined processes, it must base decisions on quantitative measures of its processes and products so that objectivity can be attained and rational decisions made. 4. Quantitative - Detailed measures of the acquisition processes, products are collected. The processes and products are quantitatively and qualitatively understood and controlled. 5. Optimizing - Continuous process improvement is empowered by quantitative feedback from the process and from piloting innovative ideas and technologies. Ultimately an organization recognizes that continual improvement (and continual change) is necessary to survive. The structure of the SA-CMM is consistent with the SW-CMM. It is a staged model with Key Process Areas (KPA's) grouped at the five maturity levels shown on the chart below (Table 1).
...the Software Engineering Institute been going strong for 19 years. The good stuff is at: http://www.sei.cmu.edu/
19 years only ? It is about 65 years that Walter Shewart, invited by Edwards Deming, has began to spread his method through a major conference which becomes a book on foundation of Quality Method. But Deming knew it would be a slow process as he said - years ago - (I translate from french because I don't have the english version) "Half a century has passed since the important publication of this book. One more half century will probably be needed before the whole thinking of Shewart will be entirely known in Education, Science and Industry." And since I have been prject manager consultant in methodology for software engineering for several clients missions, I know that it is far from being well spreaded in practice since many open big eyes when hearing some terms or concepts from Quality Management field. Organisation exists for a long time but mentality doesn't penetrate so quickly although at some point it accelarates and I think that this point has finally being reached after 10 years of stagnation. The problem is that many unexperienced quality engineers tend to have buraucratic or police control mentality instead of genuine quality improvment process mentality and generally they have poor understanding of variation concept, they generally think that quality opposes costs which is in the "western" mentality which deming totally opposed. but it's not their fault it is their teachers who tell them that as I was told myself until I met a friend of Deming who reexplained me the true spirit of Deming and Shewart approach - I admit that I found him weird on the moment - for example he said Quality Circle are useless (too much psychological bla bla bla not concrete action, he said Quality circle should be done at the end not at the begeinning) and that all Quality ISO qualification are often bullshit because they are just papers and don't imply real commitment from the firms who engaged the ISO process - although I remember he had been in the past president of an ISO quality organisation - and I really did understood what he said only a few years later . Now I know only read the founders not the "modern" ones - That's what I said also for Technical Analysis don't trust the "modern" technical analysts authors who seem to have lost the true understanding of Market's action.
...my comment was from the narrow perspective of managing DoD projects with SW people working for me in an SEI level 3 environment. It is good to know what your background is to put your posts into better context. I was unaware that this discipline had not spread to commercial practice, as I am not professionally a SW person. The reference to Shewart is much appreciated, as it was before my time. From this perspective, my caveat is the following. I suppose that you mean that trading success could benefit from the discipline of the SW CMM. However, in my work experience that methodology seriously gums up the works and is an impediment to rapid progress. In my own personal SW development I am an advocate of the "new" so-called Method-X, balls-to-the wall, to-hell-with-the-discipline SW development. I want it now, I want it cheap, and the bugs will manifest themselves soon enough in practice. Best regards. - Mike
...it is disrespectful to make fun of the elderly, especially Harry who is probably at least double my own advanced age. When Harry used "bullshit" and "ISO 9000" in the same sentence, he gained my instant respect (briefly). He needs to get up to date, however, as the latest quality BS is ISO 2000. For all I know, Harry is an ISO auditor! Any way, I avidly look forward to understanding how to apply SW CMM to trading improvement. I have already started writing those Software Critical Item Specifications and Algorithm Description Documents I so shamefully skipped when developing my current systems. Also, I will not now release code until it has gone through my quality department. BTW, if you are posting, you must not be trading. We gotcha! My excuse is that NQ needs a decongestant so that it can resume its inexorable move to Harry's next crossover. Range traders have no doubt made a trillion dollars while I am waiting for a breakout. Actually, the joke is on you. Harry needed a vile dastardly enemy, so he invented me.
I'm not really focusing about SW CMM per se, it's rather to give an idea of the ITERATION process of Shewart-Deming PDCA (Plan Do Check Action) wheel that has been deployed in SW CMM but I could take something else, it's just that I have encountered the link by searching something else on yahoo . What I mean is that the STEP by STEP philosophy or ITERATIVE PROCESS is very important in Deming's philosophy and that in software engineering this is now widely adopted whereas before the usual philosophy wasn't iterative at all: the process was supposed to be very linear. The SW CMM defines the level of each step and one could inspire by defining same kind of level in one's trading process. I haven't practiced Method-X - perhaps too new or Xtreme for me - I have practiced RUP - although I found it too bureaucratic with many papers harassment - but the philosophy of both is derived from PDCA although they don't explicitely refer to it (in fact RUP is rather a framework methodology than a method so that it can embrace Method-X as tool).
If I was double your age, you would not be a teenager yet but nearly . No I am not and have never been an ISO 9000 auditor - I only talked about the guy I invited not about me - I have been only Quality Engineer at the beginning of my career - in fact I was both quality and software engineer for food and pharmaceutical industry before being completely software engineer with my own business. I wrote a few ISO procedures but that's not what I liked to do. I first discovred Quality Control because I had organised some conferences for the group of our 5 engineering schools that's why I had the occasion to meet Deming's friend for a few days - he didn't even want to be paid for that normally he is a highly profile consultant and so very costful. I remember he was angry because the poster of the conferences I have ordered was illustrating the "seven zeros" concept (zero default, zero paper etc.) which were Bill Crosby invention : he told me that Bill Crosby - considered also as a Quality guru by the domain - was a guy he worked with at AT&T and that he invented these terms only because he had some difficulties to grasp the statistics concept and that unhappily the "seven zeros" are dangerously misunderstood by the people because they think that you must absolutley get zero defaults at any price which conducts to action that contradicts the PDCA philosophy of CONTINUOUS look for progress: the result is not the focus but the process, if the process is good then the result will be also good WITHOUT WORSENING THE COST but DIMINISHING IT. As for applying SW CMM as I said above it's not the SW CMM per se, it's rather an example of PDCA concept that is used in a discipline like software acquisition process but it could be anything else like trading, swimming or fishing . For example, if novices in trading were conscious of the necessary LONG and ITERATIVE PATH they would work instead of looking for recipes to apply. Now my intention, after I have achieved my model - although I have some other connected research to do - is to create a whole framework for trading domain adapted from Deming's philosophy. The reason is that as I will create some hedge funds and recruit traders to manage them I will have to form them and check their performance. When I look at today's methodology in finance Industry I'm rather skeptical - for example comparing performance of a trading system or a trader with buy and hold is like comparing pear and orange. I haven't began yet I intend to do so towards middle of year 2004.
...do you have examples in mind of how the iterative continuous improvement process with quality control can be applied to building a trading career? I ask in all seriousness because most of what I discover is random, like a blind hog finding an acorn. I realize you didn't mean to say SW CMM was applicable per se, but I have to take one more bitter shot at it: I don't know how you code, but I would have to decrease my productivity 20-fold to get DOWN to the 1 SLOC per hour productivity rate in my industry. Best regards. - Mike P.S. - I trust you noticed that you were labelled a "resident idiot" on another thread, along with GG and Candletrader. Quite an honor! I was deeply offended not to have been included.
Harry, I have always found your posts as some of the most valuable here. After looking through the link you provided in my last thread, http://www.elitetrader.com/vb/showthread.php?s=&threadid=22821&highlight=ramoutar I read into link, and there are four words that strike me with stunning force. Plan, Do, Check and Act. Plan- Our Personal Trading Plan: people who are religious in their faith follow some sort of code or rule. Dating back to ancient times, the Code of Hammurabi was used to instill order, discipline and a path to follow. Religious people of faith follow their book. This book (the plan), their faith (belief in the plan) and religious adherence to the faith and book (discipline) is what they believe will help reach the desired destination of their faith. While I have a global plan which covers all aspects of my trading, I also have a trading plan for each trade. My last trading plan todayâ¦go long CEPH: entry 47.45, stop 47.25, target 48.05, risk reward ratio 1:3. Do- After having complete faith in the plan I act. I cannot control what the result of this trade is going to be, that is why I have a stop. By predetermining the stop I have predetermined the risk, and I am at great peace with it. This has take a major weight and pressure from the emotional part of the trade. Looking the risk right in the eye, looking at the fear and running with the trade. The greed is also addressed by the predetermined target. This compels me to take the profit, and check or reevaluate with the money in my pocket. Check- The only I checked in this trade was the momentum. Through momentum and technical analysis I can quickly see the odds of this trade unfold. The NAZ was stomped, SPUs found no major relief and CEPH kept trading the offer through and hard. My target was imminent. I use âSCOUTSââ¦putting little bids below and near questionable support areas. None of them were hit in the ambivalent stage of the move. This told me to hold for target. Act- If the scout bids were nailed I would have hit the highest bid and closed. The positive side was moot since I was not taking the position into tomorrow. The link, which you provided, is very helpful and one can glean a lot of valuable concepts from hit. However, the author does not seem to have ever tradedâ¦.for a living at least. My definition of discipline is the fortitude to stay the course. Fortitude is the faith and belief to stick to and execute within the plan. The course is the plan. The link you provided can make it very easy for one to over intellectualize the importance of the plan and the discipline to follow it.