Search Documents

My roots span four generations. My great grandfather, Christian Frederick Fisher, along with twenty-one others was drowned in the New Australasian mine disaster at Creswick in Victoria in 1882. My grandfather, also named Christian Frederick Fisher, escaped and then followed a career of shaft sinking for large companies and operating small mines (very unprofitably according to my father) when funds allowed. My father, an excellent practical mining and mechanical engineer, was also a small miner and my early entry into the industry was, like many before me, very basic. My introductions to the business end of a No. 4 shovel, drilling with hand steel and later with jack hammers without the benefit of airlegs, feeding heavy cordwood to hungry boilers and felling trees for mine timber soon dispelled any illusions I had that mining was a bed of roses.Before it became well known I knew "Life was not meant to be easy".In a family-owned small gold mine in a remote part of Cape York Peninsula in far north Queensland I was fortunate in obtaining an excellent grounding in reef and deep alluvial mining. Operating in vertical and underlie shafts the variable ground conditions often required close timbering. Steam pumps were used underground and steam winders for hoisting.The complete absence of municipal or government infrastructure in this remote mining field threw responsibility on each mine operator to be self-sufficient, and responsible for mine development and installation of crushing and milling equipment, accommodation and water supply. Conditions, although harsh, were acceptable in the 1930s, and fostered great self-reliance.The first crushing from our Black Cat Mine in 1935 returned nearly 200 oz from 10 tons of ore. Gold was then worth $13 an ounce compared to about $400 an ounce today; but in those depression years of the 1930s our operation was profitable.This basic education not only returned dividends when confronted with difficult mining problems, but created a deep respect for the miners who develop the small mines and who continue to contribute to the industry today, often under difficult conditions.Australia and mining are synonymous. The industry has been immortalised by the famous historian, Professor Geoffrey Blainey, in his book "The Rush that Never Ended".Initial gold discoveries were concealed and exploration was frowned upon by authority as being disruptive to early rural development. But with the discovery of gold in the Bathurst area of New South Wales by Hargreaves in 1851, the rush to the gold fields began followed by an influx of migrants. The traumatic effect on industry and society that followed, created the pattern of the love-hate relationship that at times became temporarily submerged, but is so evident today.In the harsh economic climate of the 19th century with low wages and no welfare, the opportunity to wrest riches from the earth was literally "rushed" by the adventurous, the poor, the rich, people from all walks of life. Pressures developed on preFederation Governments to provide law, order and infrastructure to the rushes that have never ended.

Jan 1, 1984

By Patterson B. G

The small party of mining men who formed the original Mount Morgan Syndicate in 1882, commenced operations on the top of the mountain with what were practically nothing more than small hillside quarries, It was not until 1891 that underground stoping began, and open-cut mining therefore commenced with the commencement of operations. This hillside quarrying, however, seems to have been quite haphazard and unsystematic ; there is no trace or sign of any properly devised mining scheme. This can be seen plainly enough from an examination of old mine plans; for in less than five years after stoping was; adopted, the open-cut was down into the stoped-out ground. The following table, prepared from these old plans, shows the downward trend of the excavation in the past fiftyseven years:-

Jan 1, 1939

The Electricity Generating Authority of Thailand (EGAT) Mae Moh coal mine is an open-cut coal mine in Thailand that produces about 16 Mt of lignite annually to generate 2400 MW of electricity. The surface mine pit covers is 4 km by 7 km with mining depths ranging up to 260 m. A large pit slope is formed with the progression of the mining operation. However, massive slides in the pit slope occur often due to the existence weak planes such as faults and bedding, and the weak mechanical properties of the rocks. Hence, 200 m to 300 m boundaries of rock block including coal seams are left in front of faults in order to prevent slides and maintain the stability of the pit slope. As a result, there are significant coal reserves beneath the abandoned area along the pit slope. Moreover, the mined coal seams are very thick and their thickness range from 15 to 30 m.This paper discusses the application of multi-slicing highwall mining system with stowing to this very thick coal seam left around large pit slopes by means of numerical analysis.

Dec 6, 2010

The Illawarra Branch of the Australasian Institute of Mining and Metallurgy over recent years has established an enviable reputation for organising and conducting conferences, colloquia and symposia. And this Symposium must surely enhance that reputation. The subject of the Symposium, Seam Gas Drainage has become a matter of great import- ance to the Australian Coal Industry in a comparatively short time. Our short experience being due to our operations, until recent times anyway, not being unduly affected by gas. The Australian Industry, by virtue of its mines getting deeper and higher rates of extraction being required is now encountering the difficulties with gas which have for so long been part of the coal mining scene in many other countries. Because these gas emissions affect our working conditions by way of creating hazards and affect our production by way of bringing about cessation of operations from time to time this Symposium must be regarded as being very timely out here. I must congratulate the Branch Committee on the general set,yp :o.f the programme of papers. The classification of papers into four groups and the sequence in which the papers were presented provided an excellent flow through from introduction to finality. It is not my intention to further discuss individual papers - already they have been well discussed in this forum and as I understand it outside in the early hours of the morning,

Jan 1, 1982

By Pandey A. K, Srivastava A, Das A

Drilling and blasting constitute important activity in mining operations and account for around 20% of the total excavation cost. The utilization coefficient of energy for explosions of the fragmentation type is hardly 15 to 2035 and an enhancement in this is possible only through optimization of blast design. Designing a 'blast is a complex planning task, which for optimum results calls for matching the energy level of the explosive, rock properties and bench geometry. Optimum blast design needs a comprehensive study and an objective analysis of the intersecting elements. A cost optimised approach is often adopted with due consideration for all the cost sectors viz., drilling, blasting, loading, hauling dumping, and primary crushing, contributing to the overall miring cost. The present study deals with a performance prediction model which predicts the optimum burden for an explosive system in a given rock type to produce a certain desired fragment size. The model is framed on the inputs : shovel size, drill size, unit cost of drilling, bench dimensions, rock density, bond index, unit cost of explosive, maximum work potential. The model has been extended to compute the drilling and blasting costs for an assumed size of the bench to be blasted per round, so as to arrive at the overall cost optimization from a number of explosive systems. A comparative cost calculation with different explosive systems has been presented for a typical iron ore mine in India.

Jan 1, 1990

By D E. Langberg, J S. J va, D R. Swinbourne

Specifications on the types and levels of impurities tolerable in metals are becoming increasingly stringent, and are pushing the limits achievable with conventional refining technologies. Particularly difficult is the removal of electrochemically noble elements with low volatilities, such as copper, from more active host metals such as lead, iron and aluminium. Electrochemical refining, utilising the ability of certain solid electrolytes to selectively transport copper or other impurity cations away from the molten host metal into a separate refining phase, may provide an answer to this problem. Numerous studies on solid electrolytes used in concentration sensors have been reported, as well as a limited number on metal refining. Oxygen anion conducting solid electrolytes have been applied to the deoxidation of copper and steel while cation conductors, eg ¯ -alumina and Nasicon materials, have been used to remove lithium and sodium from aluminium. However, solid electrolytes have yet to be applied to the removal of noble elements. The behaviour of solid electrolytes under the extreme conditions of molten metal refining is not well understood. This paper will examine the potential use of solid electrolytes to remove noble metals such as copper from active host metals such as iron. The suitability, and selection criteria of available solid electrolytes for implementation in the metal refining industry in the future will be discussed. In particular, the stability of Cusicon, a copper superionic conductor potentially suitable for use in metal refining, has been assessed and the results of the experiments are presented.

Jan 1, 2000

By K McNabb, M H. Turner, P A. Jenkin

Mining of the Agnew nickel deposit commenced in 1970 using a variety of underground mining methods, though with mixed success due to the difficult ground conditions. WMC Resources Ltd acquired the operation in 1989, and commenced nickel production with an open pit. In 1995 all production reverted to the rejuvenated underground mine, using exclusively SLC (sublevel caving) methods. Currently the mine produces 1.4 Mt per annum, at an average grade of 1.74 per cent Ni. Underground mining of the Perseverance orebody has always presented significant challenges due to the difficult ground conditions in the weak and altered ultramafic rocks. The range of problems include areas of swelling minerals, intense and extremely weak shear zones, relatively high in situ stress, and in other areas very brittle rocks. Under these conditions rock mechanics input has been and remains a critical element of the mining strategy. Severe ground behaviour difficulties down to a depth of 500 m below surface prompted a review of the mining strategy for the large disseminated nickel orebody, including evaluation of alternative mining methods. However, sublevel caving was ultimately selected as the preferred method, and a re-design of the SLC cross-cut layout followed. A selection of rock mechanics investigations of interest are described, including complex virgin stress measurements, detailed ground behaviour monitoring, and the use of non-linear stress analyses to re-design the mine layout. Subsequent ground behaviour, support design and production scheduling are also described.

Jan 1, 2000

By A Richmond

"In some lode-style gold deposits the geological (lithological) uncertainty may have similar economic significance as the grade uncertainty. In the absence of precise knowledge of the deposit geology, geological models of these types of deposits that are used to constrain spatially the estimated (or simulated) grade values may influence significantly the contained metal. Methods for assessing geological uncertainty quantitatively are rarely considered due to their perceived inability to consider exhaustive geological interpretations, and resulting models that often appear geologically unrealistic. Furthermore, should the lodes have been subjected to folding, or where the mineralisation continuity does not follow a planar orientation along strike or down dip, standard methods of variography and interpolation using linear vectors in space often do not provide a good representation of continuity. An alternative approach is to relate spatial samples from adjacent drill holes using an unfolding plane.The paper demonstrates conditional simulation of the geometry and gold grades of a folded lode-style deposit where:simulation of lode geometry considers explicitly the geologistÆs interpretation, andsimulation of gold grade values that consider explicitly the folded nature of the lodes.The results are contrasted to the traditional approach of linear vectors in real space for a single lode interpretation.CITATION:Richmond, A, 2012. Conditional simulation of a folded lode-style gold deposit, in Proceedings Narrow Vein Mining 2012 , pp 149-154 (The Australasian Institute of Mining and Metallurgy: Melbourne)."

Mar 26, 2012

By R Rodriguez, S Gumo

The Lihir Gold Mine, in Papua New Guinea, is a large open pit operation within an active geothermal system inside a collapsed volcanic crater. This unique setting has a big impact on the mine operation because of the occurrence of geothermal hazards. The system for managing geothermal hazards that has been developed by Lihir Gold Mine is a pioneering effort that can be applied to future mine discoveries in a similar environment. The controls are integral to the mining operations and allow mining to proceed whereas without them mining could not be considered. Truck and shovel productivities in conventional open pit operations are constrained by diggability, equipment availabilities and utilisation. At Lihir, additional constraints are in the form of 100 m exclusion zones around working shovels and 50 m exclusion zones around working drills. Given that the mine has seven large digging units and six drills, this amounts to a lot of area that excludes the presence of people on the ground. So how are the necessary work like grade control, survey, blasthole loading, and in-pit maintenance performed safely to achieve production targets? The answer is in the highly specialised procedures that have been developed that utilise geological, geothermal and geotechnical knowledge to predict problem areas and also deal with the organisation of movement in the pit for people and mining equipment. This paper will detail how scheduled material movement can be maintained despite the challenging conditions faced at Lihir Gold Mine.

Jan 1, 2007

By Singh T, Singh D. P

India has limited resources of coking coal. Most of the coking coal seams are located in Jharia Coalfield. These seams were worked in the past by the bord and pillar system in a haphazard manner, which has resulted in honeycombs of old working. Most of such areas are under fire. In order to meet the demand of coking coal in the 21st century, it is essential that the coal locked due to old underground workings is recovered fully by conducting opencast mining to an ultimate depth of 500 m. The major problem envisaged in conducting opencast mining in such areas will be the stability of pit slopes over voids and the behaviour of subsided material in the pit slopes. An attempt has been made to simulate this complex situation from a specific section of the rockmass of a mine in an equivalent material (EM) model. The old workings existing in upper coal seam (V) were simulated by developing staggered bord and pillar workings in two sections and depillaring them. Opencast mining was conducted for recovery of coal locked due to irregular working in coal seam V. The slope instability caused by intersection of the slope with the void, during frontal working (by shovel-dumper combination) was observed in various situations. In this paper, the technique adopted for simulation of a 125 m thick section of rockmass in an EM model is outlined. Adverse effects of staggered workings and

Jan 1, 1988

By Adeloju S, McQuade CV, Riley SJ

Metal contamination of sediments can be an issue for minesites discharging water. Water sampling of receiving waters is frequently undertaken. Sampling and analysis of sediment is less common. The variability of receiving environments makes the formulation of a generic sediment sampling strategy virtually impossible. Each situation needs to be carefully assessed before sampling is undertaken. This paper outlines the approach taken for one such survey.The main factors that need to be considered in a survey of surficial sediments are sample representativeness and sample variability. A simple data set from one site within a regional survey is used to illustrate the importance of these factors. The metal content of sediment is shown to vary significantly over small spatial scales.The variability associated with the spatial distribution of samples within a site is shown to be the most significant source of variability in the study. This variability could bias the results of a study if not planned for in the sampling methodology. Taking multiple samples from within each site (sub-site samples) and combining the data gives a more representative indicator of overall conditions than a single sample. The optimum number of sub-site samples to be taken from each sampling site was found to be 15.It is shown that a composite of sub-site samples can give a good indication of the average site metal content, while considerably reducing the sample preparation and analysis effort.

Jan 1, 1998

By Dight P. M

The challenge of narrow vein mining is to produce ore with as little dilution as possible, at the least cost -commensurate with the increasing degrees of safety being demanded for mine workers by the modem 'duty of care' legislation. Geotechnical factors play an important, if not overriding, role in these areas. The geotechnical factors cover the complex interaction over a period of time, between the in situ and induced rock stresses, rock structures, rock mass strength and the cable, rock bolt and other ground support members. In the Archaean schist belt areas where many of our narrow vein mines occur the pre-mining in situ rock stresses are commonly significantly higher than the overburden stresses. These are concentrated in areas beneath previous open pits and may be highly concentrated in the pillars and abutments of stopes on narrow veins. The presence of high in situ stresses may be suspected from the appearance of drill cores and underground development long before they become sufficiently concentrated by mining to cause damage or bursts. These initial indications are discussed and modem low cost methods of stress measurement are described. Practical and quick methods of analysing the stress redistribution around scopes and other underground excavations are now available and guidelines for interpreting them in terms of rockmass characteristics, stope stability and support requirements are developed with the object of increasing safety, whilst reducing overall costs and dilution.

Jan 1, 1993

The Indonesian province of Papua is a remote jungle-clad region covering the western half of the mountainous island of New Guinea. Despite being relatively under-explored to this day,Papua is home to the greatest mineral success story of the 20th century. The Grasberg/Ertsberg minerals district atop the central dividing range contains the worldÆs largest reserve of both copper and gold (Freeport-McMoRan Copper and Gold, 2002). With large mines at Ok Tedi,Por gera and Lihir,and many well-delineated resources such as Frieda River,Wafi and Hidden Valley, neighbouring Papua New Guinea,the eastern half of New Guinea has enjoyed sustained exploration campaigns for gold and copper since the mid 1960s. The combined Grasberg and PNG exploration successes can be measured in terms of ounces of gold produced over the last decade versus available landmass,conf irming the island of New Guinea and surrounding islands as a world-class destination for gold explorers. Yet despite this obvious regional prospectivity and the profitability of the Grasberg operation, a sustained exploration effort in the Papua province in Indonesia has not been achieved. Periodic political constraints, poor infrastructure, high costs, operational challenges, social and environmental concerns and lack of immediate exploration success often impact exploration investment decisions. Accordingly a lesser number of mineral occurrences have been documented in Papua compared to those seen along tectonic strike in geologically similar Papua New Guinea.

Jan 1, 2004

By V K. Singh, G Sural

The problem of spontaneous heating/fire is very common in coal mines all over the world. The coal industry faces many problems day by day due to the occurrence of spontaneous heating in the open pit coal mines. Presently, about 70 per cent of total coal production in Indian mines is being carried out by open pit mining. Fire in coal, particularly in open pit mines not only causes irreparable loss of natural wealth but also damages the surface structures and pollutes the environment. It is well known that 75 per cent of the fires occur due to the spontaneous combustion of coal. Spontaneous heating is a very common occurrence in coal mines and is a major factor contributing to mine fire. The liability of coal to spontaneous heating causes many serious problems before, during and after mining. In opencast mines, coal immediately oxidises and catches fire due to the intrinsic characteristics of coal, such as low rank, high moisture, high volatile matter, presence of sulfur in the form of pyrites, low crossing point temperature (CPT) and ignition point temperature (IPT) value and less incubation period. In opencast mines, when the coal benches are left idle for a longer time, heat accumulation takes place in favourable conditions and sometimes leads to fire. The purpose of this paper is to present the different successful case studies regarding the safety management of open pit coal mines from occurrences of spontaneous heating/fire.

Jan 1, 2007

THE irregularity of the Mount Morgan ore body presents problems other than those for mining only. For irregularity of outline is coupled with irregularity of grade, and the ore varies so rapidly from profitable to non-profitable, and from highly siliceous to highly pyritic, that the task of estimating quantities and values of ore reserves is manifestly rendered a difficult one, while the actual disposal Of the broken ore demands constant supervision.Four classes of ore are handled. The oxidized ore, cheaply mined and cheaply treated, is obtained by quarrying from the top of the mountain. Below the 315-ft. mine level (measured from the original summit) sulphide ore comes in, and this is divided into siliceous sulphide ore (too siliceous for economic smelting), sulphide ore suitable for smelting, and ore too poor either to be smelted or to be treated by wet process with the siliceous ore. It is with these three latter descriptions of ore that this paper principally deals.Although the output from most stopes in the mine may usually be presumed beforehand to be either smelting or wet-treatment ore...

Jan 1, 1910

The ~2500 km long Kermadec-Tonga arc northeast ofNew Zealand is the longest continuous intraoceanic arc in the world. At least 94 volcanic centres along the arc are submarine, most (87 per cent) within the Kermadec arc (de Ronde et al, 2003a). Volcanic rocks range from basalt to rhyodacite with trace element and isotopic data indicating significant magma source heterogeneity both along and across the arc.   The entire ~1300 km of the Kermadec arc has now been surveyed for hydrothermal emissions (de Ronde et al, 2004). First-stage surveys utilised surface vessels. In February 1996, the first sulfide samples from the southern Kermadec arc were dredged from the Brothers and Rumble II West caldera volcanoes (R/V Tangaroa cruise TAN96/03; Wright et al, 1998; de Ronde et al, 2003b). In September 1998, low-temperature, diffuse venting was discovered at Monowai, Rumble III and Clark volcanoes, and high-temperature, black smoker venting was found within Brothers caldera (R/V Sonne cruise SO-135; Stoffers et al, 1999). The NZAPLUME cruise of March 1999 surveyed 13 volcanic centres (each mainly comprised of single, large volcanoes) of the southern Kermadec arc and documented hydrothermal emissions above seven of them (de Ronde et al, 2001; Baker et al, 2003; Massoth et al, 2003). In July 2001, additional mineralised and hydrothermally altered samples were dredged from Brothers volcano (R/V Tangaroa cruise TAN01/07). The NZAPLUME II cruise of May 2002 confirmed venting was still occurring at all the known hydrothermal sites, and that ten active vent sites occurred along the mid-part of the arc, between Brothers and Raoul Island (de Ronde et al, 2002). In September-October 2004, NZAPLUME III surveyed the northern-most part of the Kermadec arc where another nine volcanic centres are hydrothermally active.   Second-stage exploration of the Kermadec arc hydrothermal systems utilised deep submergence vehicles. In October-November 2004, the Japanese submersible Shinkai 6500 made four dives on Brothers. Samples recovered include vent fluids, chimneys, altered rocks and various animals and microbes. In April-May 2005, a more extensive expedition using the American submersible Pisces V made 23 dives on nine different volcanic centres, eight of them for the first time.   Extensive plume surveys show Brothers volcano is host to the most hydrothermally active sites of all those known along the Kermadec arc. In total, nine dives have been made on Brothers, six on the NW caldera site and three on the cone site. The NW caldera vent site is a long-term hydrothermal system that is today dominated by evolved seawater but has had episodic injections of magmatic fluid. The cone site is a nascent magmatic-hydrothermal system where crack zones localise upwelling acidic waters (de Ronde et al, 2005). A SE caldera site represents the main upflow of a relatively well-established magmatic-hydrothermal system on the seafloor where sulfide-rich chimneys are extant. Each of these different vent sites represent diverse parts of an evolving hydrothermal system, any one of which may be typical of submarine volcanic arcs.

Jan 1, 2005

By D P. Blair

There are two major aspects to wall control blasting – the potential for any blasthole to cause wall damage and the potential for a delayed sequence of blastholes to cause damage. The damage potential of individual blastholes is governed by standoff distance and charge weight. However, for locations sufficiently close to a highwall, dynamic finite element models and field measurements show that there is also a critical nature to those blastholes very close to the wall, over and above standoff distance. At many mine sites, the charge weight/standoff-distance is set by trial and error. Thus, a significant aspect of the present work is the determination of dynamic stress radiating from a blasthole. Unfortunately, the modelling and measurement of this dynamic stress is still faced with many difficulties. Typical models have questionable aspects, which casts serious doubt over their predictions, and it is shown that the Hustrulid Bar experimental technique does not correctly account for stress wave attenuation in extended media. In an attempt to solve the dynamic stress problem, the present work describes a model based on an exact solution under the assumption of viscoelastic material. The model takes full account of the rock mass elastic/viscoelastic properties, the explosive type, geometry and primer location and predicts a dynamic stress for direct comparison with the unconfined compressive strength of the local rock mass. This model predicts the standoff distance required to avoid damage due to a single blasthole. The behaviour of a delayed sequence of blastholes is analysed using an upgraded waveform superposition model, covering a range of initiation delay combinations, which predicts the peak particle motion (PPM) induced in the wall for each combination. This approach shows that there is no need to fire dedicated trim shots close to walls; all that is required is a carefully modified production shot. The mechanism of vibration screening is discussed, and its use in a superposition model highlights the potential of reverse firing (shield blasting) to reduce wall PPM. A conceptual model is also introduced to give a further insight to vibration screening and damage (via ground heave). This model predicts that a centre-lift blast will have a central heave profile and that a reverse-fired shot will produce a smaller power trough than that produced by a standard shot. It is demonstrated that the mechanism of presplit formation is not well understood; nevertheless, experimental evidence shows that presplits can be an effective method for reducing wall vibrations.CITATION:Blair, D P, 2015. Wall control blasting, in Proceedings 11th International Symposium on Rock Fragmentation by Blasting, pp 13–26 (The Australasian Institute of Mining and Metallurgy: Melbourne).

Aug 24, 2015

From the ventilation aspect, there is a basic difference between advancing and ezreating longwall layouts. Particular rroblems with retreating, using differing mining methods and layouts include high gas emissions, spontaneous combustion and -radients. There is an acceptable method of crntrolling the environment at the return end a retreating longwall face without the use a multiple heading system. Incorporated in the design is the facility to practice xtensive methane drainage, thus ensuring a ntrol of the rate of methane emission into the main roadways. Case studies highlight varying mining editions under which the proposed system of Ctised.

Jan 1, 1982

By J Lea, P Hatherly, J L. Edwards

Coal exploration has traditionally involved the acquisition of data to enable a resource to be quantified and qualified to the extent that its economic potential can be determined. This process commonly relied on: drilling to determine the depth and thickness of coal seams, broad characterisation of the overburden and interburden, and to provide cores for coal quality analyses; basic wireline logging to cross-check the depth and thickness of coal seams and provide some indication of rock mass strength; and basic surface geophysics and remote sensing to identify major geological structures. In recent years, the adoption of the JORC Code as a basis for the reporting of resources and reserves has placed a greater emphasis on the need for detailed, reproducible data sets that not only satisfy the traditional requirements of mine planners, but also satisfy the governance requirements of corporate and financial institutions. Similarly, the development of sophisticated numerical modelling techniques for use in both open cut and underground mine design, has seen a greater demand for precise geotechnical data, in situ stress characterisation and hydrological information as routine input to the design process. These requirements, coupled with the prevailing trend toward high capital, relatively inflexible mining techniques less tolerant of geological surprises, and the need for consideration of environmental issues such as groundwater impacts and surface subsidence, requires that modern exploration programs be multifaceted, and designed to commence capturing data relevant to ædownstream issuesÆ even during the initial phases of exploration planning.

Jan 1, 2009

In order to realise seafloor massive sulfide (SMS) mining, the environmental baseline survey around targeted SMS mining areas and the creation of an environmental assessment technique suitable for quantifying the impacts of ocean floor mining are necessary. The targeted areas generally host large chemosynthetic communities in association with active hydrothermal venting. The ecosystem structure around the chemosynthetic communities has not been clarified yet in macro-scale or quantitatively. The systematic approach and plan for the baseline survey and the environmental assessmentÆs basic studies are introduced. An EXTENDED ABSTRACT is available for download. A full-length paper was not prepared for this presentation.

Jan 1, 2008