Levels of Selection, Logical Schemes, Selfish Genes, and Misleading Memes

AppId is over the quota
AppId is over the quota

Both Nature and Science are currently celebrating the 100th anniversary of the birth of an icon of logic, computer science, and mathematical biology: Alan Turing.  In reading Andrew Hodges’s spectacular biography of Turing (1983) many years ago I came to appreciate that the subject of the book was both a deeply creative and extraordinarily rigorous thinker.  Although Turing is known for seminal achievements in mathematical logic and computer science, his most directly practical and immediately consequential contribution was his facilitation of the Allied cause in World War II through his guidance of the effort to break the Nazi military code.  This effort called primarily on his prodigious talents for far-reaching inference and it was in reading about this effort that I was prompted to consider a concept that might be called “maximum deduction.”  Turing and his able colleagues needed to make every possible deductive inference (or at least very close to every possible inference) supported by the available data on German military communications in order to solve a problem of immense and immediate impact (the saving of Allied ships from devastating German submarine attacks).

In reading Samir Okasha’s thorough and insightful guide to the theoretical debates about multi-level selection, Evolution and the Levels of Selection (2006), I am reminded of Turing’s logical rigor. Like Turing, Okasha possesses the ability to fully explore the implications of an intellectual position.  Of similar value, he makes key distinctions that elude or at least receive inadequate attention from others and fairly assesses alternative conceptual schemes or theoretical approaches.  For example, his examination of the relative merits of the Price equation versus what he refers to as the “contextual analysis” for assessing and partitioning selection in differing evolutionary scenarios reveals that each has important advantages as well as significant weaknesses.  Much to his credit, he does not seek a neat but oversimplified and misleading conclusion.  The figures are simple but effective and substantially aid the exposition.

I cannot attempt to summarize all of the arguments in the book of about 240 pages because Okasha’s arguments are of sufficient intricacy and subtlety that it would be nearly impossible to substantially compress them without causing serious distortions in the reasoning.  Therefore, I will just note the topics addressed and remark on a limited number of particularly interesting points.

Okasha begins the book by introducing and characterizing the levels-of-selection problem and explicating the essence of natural selection in abstract formal terms.  He then addresses the distinction between the synchronic and diachronic perspectives, where the former deals with the hierarchical organization of the living world (e.g. cells, multicellular organisms, communities of multicellular organisms) such as it is and the latter is concerned with how the hierarchy arose.  Next, the author introduces and explains the interpretation of the equation formulated by George Price forty years ago to describe in mathematical terms the evolution of a population from one generation to the next.  He also delves into the sometimes-consequential differences between statistical and causal decompositions of changes in organismal characters across generation.  This chapter ends with an interesting discussion of the connections between the Price equation and the formal conditions for evolution promulgated more than forty years ago by the eminent population geneticist, Richard Lewontin, which were initially described at the beginning of the chapter.

The second chapter explains the fundamentals of multi-level selection, including explorations of life cycles, relevant definitions of fitness, and the distinction between multi-level selection 1 (MLS1) and multi-level selection 2 (MLS2).  For MLS1, what Okasha calls the ‘focal’ level is concerned with the number of offspring, in the next generation, of the particles that constitute a collective or group and for MLS2 the ‘focal’ level is concerned with the number of offspring groups in the succeeding generation.  Another key distinction that Okasha addresses is that between aggregate and emergent properties of collectives.  Okasha then tackles heritability and how the concept differs for MLS1 and MLS2 and includes a discussion of how the Price equation can be applied for the two types of multi-level selection.

Chapter three focuses on notions of causality and deals with the fairly subtle notion of cross-level by-products in which apparent selection on one level can in fact result from selection at another level.  In this portion of the book, the author introduces contextual analysis, which relies on linear regression models, and compares it to the approach associated with the Price equation.

What the author describes as philosophical issues take up the fourth chapter.  The section sub-headings will give a sense of the subject matter being addressed: emergence and additivity, screening off and the levels of selection, realism versus pluralism about the levels of selection, and reductionism.

Chapter five is entitled “The Gene’s-Eye View and its Discontents.”  After tracing the gene-centered perspective back to R. A. Fisher and reviewing the contributions of individuals such as W. D. Hamilton, G. C. Williams, and Richard Dawkins, Okasha makes the critical distinction between a gene’s-eye view of evolution and genic selection.  In this context, Okasha notes what he believes to be a shift in position by Dawkins.  Next the author discusses outlaw genes or selfish genetic elements (SGEs).  These DNA sequences manage to be transmitted at increased frequencies into the gametes (the phenomenon of meiotic drive or segregation distortion) and, therefore, into the next generation thereby exhibiting increased fitnesses relative to non-SGEs.  Thus, I would suggest that selfishness is a quantitative not a qualitative trait.  The Price equation and contextual analysis are then compared as to how these two approaches account for the behavior of SGEs.  Okasha then demonstrates why the gene-centered perspective is not, as sometimes claimed, a completely general way to account for any evolutionary scenario, e.g., when dealing with non-genetic inheritance, plants that produce vegetative entities that are often genetically chimeric (i.e., ramets), and insect colonies founded by multiple queens or multiply-mated queens.  The genic perspective also is less obviously successful when non-additive interactions between genes are present, which is reasonably common.  An interesting point that Okasha makes is that whenever SGEs arise, there is likely to be selection on the unlinked ‘law abiding’ genetic elements to suppress the ‘cheaters’ since SGEs typically enhance their own fitnesses at the cost of diminishing the fitness of the organism with respect to which they may reasonably regarded as parasites of a sort.

The sixth chapter addresses the still active and evolving controversy or group selection as of 2006.  Okasha provides historical background, discusses the distinction between MLS1 and MLS2 in the context of the controversy, explores how ideas about kin selection, reciprocal altruism, and evolutionary game theory feature in the debates, and describes the roles of a number of other concepts in the key disagreements in the literature.

The final two chapters address macroevolutionary issues that may be less obviously relevant to those focused on the relevance of evolution to medicine.  Therefore, I will refrain from a detailed description of the content of these sections and just note an insight offered therein. Whenever there is a major evolutionary transition, as from individual genes to whole genomes or single-celled to multi-celled organisms, there must be selection against within-group conflict and selfishness of the ‘lower-level’ units and this selection must be effective for the more-complex level of the biological hierarchy to be successfully established.  Thus, one consequence of relentless competition is cooperation and all genes are not, as Dawkins suggested early in The Selfish Gene (1976, 1989), ruthlessly selfish unless ruthless selfishness embodies some measure of cooperativeness.

References

Hodges, A. Alan Turing: The Enigma. A Touchstone Book, Simon & Schuster, Inc., New York, 1983.

Okasha, Samir. Evolution and the Levels of Selection. Oxford University press, 2006.

Dawkins, R. The Selfish Gene. Oxford University Press, Oxford, 1976, 1989 p. 2.

Tags: additive characters, Alan Turing, causal decomposition, cells, computer science, contextual analysis, cooperation, cross-level by-products, diachronic, emergent characters, evolution, evolutionary game theory, focal level, G. C. Williams, gametes, genes, gene’s-eye perspective, genic selection, George Price, hierarchical organization, kin selection, levels-of-selection controversy, logic, macroevolution, mathematics, maximum deduction, meiotic drive, military code, multi-cellular organisms, multi-level selection 1, multi-level selection 2, pluralism, Price equation, R. A. Fisher, realism, reciprocal altruism, reductionism, Richard Dawkins, Richard Lewontin, selfish genetic element, statistical decomposition, synchronic, W. D. Hamilton

View the original article here

Levels of Selection, Logical Schemes, Selfish Genes, and Misleading Memes

AppId is over the quota
AppId is over the quota

Both Nature and Science are currently celebrating the 100th anniversary of the birth of an icon of logic, computer science, and mathematical biology: Alan Turing.  In reading Andrew Hodges’s spectacular biography of Turing (1983) many years ago I came to appreciate that the subject of the book was both a deeply creative and extraordinarily rigorous thinker.  Although Turing is known for seminal achievements in mathematical logic and computer science, his most directly practical and immediately consequential contribution was his facilitation of the Allied cause in World War II through his guidance of the effort to break the Nazi military code.  This effort called primarily on his prodigious talents for far-reaching inference and it was in reading about this effort that I was prompted to consider a concept that might be called “maximum deduction.”  Turing and his able colleagues needed to make every possible deductive inference (or at least very close to every possible inference) supported by the available data on German military communications in order to solve a problem of immense and immediate impact (the saving of Allied ships from devastating German submarine attacks).

In reading Samir Okasha’s thorough and insightful guide to the theoretical debates about multi-level selection, Evolution and the Levels of Selection (2006), I am reminded of Turing’s logical rigor. Like Turing, Okasha possesses the ability to fully explore the implications of an intellectual position.  Of similar value, he makes key distinctions that elude or at least receive inadequate attention from others and fairly assesses alternative conceptual schemes or theoretical approaches.  For example, his examination of the relative merits of the Price equation versus what he refers to as the “contextual analysis” for assessing and partitioning selection in differing evolutionary scenarios reveals that each has important advantages as well as significant weaknesses.  Much to his credit, he does not seek a neat but oversimplified and misleading conclusion.  The figures are simple but effective and substantially aid the exposition.

I cannot attempt to summarize all of the arguments in the book of about 240 pages because Okasha’s arguments are of sufficient intricacy and subtlety that it would be nearly impossible to substantially compress them without causing serious distortions in the reasoning.  Therefore, I will just note the topics addressed and remark on a limited number of particularly interesting points.

Okasha begins the book by introducing and characterizing the levels-of-selection problem and explicating the essence of natural selection in abstract formal terms.  He then addresses the distinction between the synchronic and diachronic perspectives, where the former deals with the hierarchical organization of the living world (e.g. cells, multicellular organisms, communities of multicellular organisms) such as it is and the latter is concerned with how the hierarchy arose.  Next, the author introduces and explains the interpretation of the equation formulated by George Price forty years ago to describe in mathematical terms the evolution of a population from one generation to the next.  He also delves into the sometimes-consequential differences between statistical and causal decompositions of changes in organismal characters across generation.  This chapter ends with an interesting discussion of the connections between the Price equation and the formal conditions for evolution promulgated more than forty years ago by the eminent population geneticist, Richard Lewontin, which were initially described at the beginning of the chapter.

The second chapter explains the fundamentals of multi-level selection, including explorations of life cycles, relevant definitions of fitness, and the distinction between multi-level selection 1 (MLS1) and multi-level selection 2 (MLS2).  For MLS1, what Okasha calls the ‘focal’ level is concerned with the number of offspring, in the next generation, of the particles that constitute a collective or group and for MLS2 the ‘focal’ level is concerned with the number of offspring groups in the succeeding generation.  Another key distinction that Okasha addresses is that between aggregate and emergent properties of collectives.  Okasha then tackles heritability and how the concept differs for MLS1 and MLS2 and includes a discussion of how the Price equation can be applied for the two types of multi-level selection.

Chapter three focuses on notions of causality and deals with the fairly subtle notion of cross-level by-products in which apparent selection on one level can in fact result from selection at another level.  In this portion of the book, the author introduces contextual analysis, which relies on linear regression models, and compares it to the approach associated with the Price equation.

What the author describes as philosophical issues take up the fourth chapter.  The section sub-headings will give a sense of the subject matter being addressed: emergence and additivity, screening off and the levels of selection, realism versus pluralism about the levels of selection, and reductionism.

Chapter five is entitled “The Gene’s-Eye View and its Discontents.”  After tracing the gene-centered perspective back to R. A. Fisher and reviewing the contributions of individuals such as W. D. Hamilton, G. C. Williams, and Richard Dawkins, Okasha makes the critical distinction between a gene’s-eye view of evolution and genic selection.  In this context, Okasha notes what he believes to be a shift in position by Dawkins.  Next the author discusses outlaw genes or selfish genetic elements (SGEs).  These DNA sequences manage to be transmitted at increased frequencies into the gametes (the phenomenon of meiotic drive or segregation distortion) and, therefore, into the next generation thereby exhibiting increased fitnesses relative to non-SGEs.  Thus, I would suggest that selfishness is a quantitative not a qualitative trait.  The Price equation and contextual analysis are then compared as to how these two approaches account for the behavior of SGEs.  Okasha then demonstrates why the gene-centered perspective is not, as sometimes claimed, a completely general way to account for any evolutionary scenario, e.g., when dealing with non-genetic inheritance, plants that produce vegetative entities that are often genetically chimeric (i.e., ramets), and insect colonies founded by multiple queens or multiply-mated queens.  The genic perspective also is less obviously successful when non-additive interactions between genes are present, which is reasonably common.  An interesting point that Okasha makes is that whenever SGEs arise, there is likely to be selection on the unlinked ‘law abiding’ genetic elements to suppress the ‘cheaters’ since SGEs typically enhance their own fitnesses at the cost of diminishing the fitness of the organism with respect to which they may reasonably regarded as parasites of a sort.

The sixth chapter addresses the still active and evolving controversy or group selection as of 2006.  Okasha provides historical background, discusses the distinction between MLS1 and MLS2 in the context of the controversy, explores how ideas about kin selection, reciprocal altruism, and evolutionary game theory feature in the debates, and describes the roles of a number of other concepts in the key disagreements in the literature.

The final two chapters address macroevolutionary issues that may be less obviously relevant to those focused on the relevance of evolution to medicine.  Therefore, I will refrain from a detailed description of the content of these sections and just note an insight offered therein. Whenever there is a major evolutionary transition, as from individual genes to whole genomes or single-celled to multi-celled organisms, there must be selection against within-group conflict and selfishness of the ‘lower-level’ units and this selection must be effective for the more-complex level of the biological hierarchy to be successfully established.  Thus, one consequence of relentless competition is cooperation and all genes are not, as Dawkins suggested early in The Selfish Gene (1976, 1989), ruthlessly selfish unless ruthless selfishness embodies some measure of cooperativeness.

References

Hodges, A. Alan Turing: The Enigma. A Touchstone Book, Simon & Schuster, Inc., New York, 1983.

Okasha, Samir. Evolution and the Levels of Selection. Oxford University press, 2006.

Dawkins, R. The Selfish Gene. Oxford University Press, Oxford, 1976, 1989 p. 2.

Tags: additive characters, Alan Turing, causal decomposition, cells, computer science, contextual analysis, cooperation, cross-level by-products, diachronic, emergent characters, evolution, evolutionary game theory, focal level, G. C. Williams, gametes, genes, gene’s-eye perspective, genic selection, George Price, hierarchical organization, kin selection, levels-of-selection controversy, logic, macroevolution, mathematics, maximum deduction, meiotic drive, military code, multi-cellular organisms, multi-level selection 1, multi-level selection 2, pluralism, Price equation, R. A. Fisher, realism, reciprocal altruism, reductionism, Richard Dawkins, Richard Lewontin, selfish genetic element, statistical decomposition, synchronic, W. D. Hamilton

View the original article here

Levels of Selection, Logical Schemes, Selfish Genes, and Misleading Memes

AppId is over the quota
AppId is over the quota

Both Nature and Science are currently celebrating the 100th anniversary of the birth of an icon of logic, computer science, and mathematical biology: Alan Turing.  In reading Andrew Hodges’s spectacular biography of Turing (1983) many years ago I came to appreciate that the subject of the book was both a deeply creative and extraordinarily rigorous thinker.  Although Turing is known for seminal achievements in mathematical logic and computer science, his most directly practical and immediately consequential contribution was his facilitation of the Allied cause in World War II through his guidance of the effort to break the Nazi military code.  This effort called primarily on his prodigious talents for far-reaching inference and it was in reading about this effort that I was prompted to consider a concept that might be called “maximum deduction.”  Turing and his able colleagues needed to make every possible deductive inference (or at least very close to every possible inference) supported by the available data on German military communications in order to solve a problem of immense and immediate impact (the saving of Allied ships from devastating German submarine attacks).

In reading Samir Okasha’s thorough and insightful guide to the theoretical debates about multi-level selection, Evolution and the Levels of Selection (2006), I am reminded of Turing’s logical rigor. Like Turing, Okasha possesses the ability to fully explore the implications of an intellectual position.  Of similar value, he makes key distinctions that elude or at least receive inadequate attention from others and fairly assesses alternative conceptual schemes or theoretical approaches.  For example, his examination of the relative merits of the Price equation versus what he refers to as the “contextual analysis” for assessing and partitioning selection in differing evolutionary scenarios reveals that each has important advantages as well as significant weaknesses.  Much to his credit, he does not seek a neat but oversimplified and misleading conclusion.  The figures are simple but effective and substantially aid the exposition.

I cannot attempt to summarize all of the arguments in the book of about 240 pages because Okasha’s arguments are of sufficient intricacy and subtlety that it would be nearly impossible to substantially compress them without causing serious distortions in the reasoning.  Therefore, I will just note the topics addressed and remark on a limited number of particularly interesting points.

Okasha begins the book by introducing and characterizing the levels-of-selection problem and explicating the essence of natural selection in abstract formal terms.  He then addresses the distinction between the synchronic and diachronic perspectives, where the former deals with the hierarchical organization of the living world (e.g. cells, multicellular organisms, communities of multicellular organisms) such as it is and the latter is concerned with how the hierarchy arose.  Next, the author introduces and explains the interpretation of the equation formulated by George Price forty years ago to describe in mathematical terms the evolution of a population from one generation to the next.  He also delves into the sometimes-consequential differences between statistical and causal decompositions of changes in organismal characters across generation.  This chapter ends with an interesting discussion of the connections between the Price equation and the formal conditions for evolution promulgated more than forty years ago by the eminent population geneticist, Richard Lewontin, which were initially described at the beginning of the chapter.

The second chapter explains the fundamentals of multi-level selection, including explorations of life cycles, relevant definitions of fitness, and the distinction between multi-level selection 1 (MLS1) and multi-level selection 2 (MLS2).  For MLS1, what Okasha calls the ‘focal’ level is concerned with the number of offspring, in the next generation, of the particles that constitute a collective or group and for MLS2 the ‘focal’ level is concerned with the number of offspring groups in the succeeding generation.  Another key distinction that Okasha addresses is that between aggregate and emergent properties of collectives.  Okasha then tackles heritability and how the concept differs for MLS1 and MLS2 and includes a discussion of how the Price equation can be applied for the two types of multi-level selection.

Chapter three focuses on notions of causality and deals with the fairly subtle notion of cross-level by-products in which apparent selection on one level can in fact result from selection at another level.  In this portion of the book, the author introduces contextual analysis, which relies on linear regression models, and compares it to the approach associated with the Price equation.

What the author describes as philosophical issues take up the fourth chapter.  The section sub-headings will give a sense of the subject matter being addressed: emergence and additivity, screening off and the levels of selection, realism versus pluralism about the levels of selection, and reductionism.

Chapter five is entitled “The Gene’s-Eye View and its Discontents.”  After tracing the gene-centered perspective back to R. A. Fisher and reviewing the contributions of individuals such as W. D. Hamilton, G. C. Williams, and Richard Dawkins, Okasha makes the critical distinction between a gene’s-eye view of evolution and genic selection.  In this context, Okasha notes what he believes to be a shift in position by Dawkins.  Next the author discusses outlaw genes or selfish genetic elements (SGEs).  These DNA sequences manage to be transmitted at increased frequencies into the gametes (the phenomenon of meiotic drive or segregation distortion) and, therefore, into the next generation thereby exhibiting increased fitnesses relative to non-SGEs.  Thus, I would suggest that selfishness is a quantitative not a qualitative trait.  The Price equation and contextual analysis are then compared as to how these two approaches account for the behavior of SGEs.  Okasha then demonstrates why the gene-centered perspective is not, as sometimes claimed, a completely general way to account for any evolutionary scenario, e.g., when dealing with non-genetic inheritance, plants that produce vegetative entities that are often genetically chimeric (i.e., ramets), and insect colonies founded by multiple queens or multiply-mated queens.  The genic perspective also is less obviously successful when non-additive interactions between genes are present, which is reasonably common.  An interesting point that Okasha makes is that whenever SGEs arise, there is likely to be selection on the unlinked ‘law abiding’ genetic elements to suppress the ‘cheaters’ since SGEs typically enhance their own fitnesses at the cost of diminishing the fitness of the organism with respect to which they may reasonably regarded as parasites of a sort.

The sixth chapter addresses the still active and evolving controversy or group selection as of 2006.  Okasha provides historical background, discusses the distinction between MLS1 and MLS2 in the context of the controversy, explores how ideas about kin selection, reciprocal altruism, and evolutionary game theory feature in the debates, and describes the roles of a number of other concepts in the key disagreements in the literature.

The final two chapters address macroevolutionary issues that may be less obviously relevant to those focused on the relevance of evolution to medicine.  Therefore, I will refrain from a detailed description of the content of these sections and just note an insight offered therein. Whenever there is a major evolutionary transition, as from individual genes to whole genomes or single-celled to multi-celled organisms, there must be selection against within-group conflict and selfishness of the ‘lower-level’ units and this selection must be effective for the more-complex level of the biological hierarchy to be successfully established.  Thus, one consequence of relentless competition is cooperation and all genes are not, as Dawkins suggested early in The Selfish Gene (1976, 1989), ruthlessly selfish unless ruthless selfishness embodies some measure of cooperativeness.

References

Hodges, A. Alan Turing: The Enigma. A Touchstone Book, Simon & Schuster, Inc., New York, 1983.

Okasha, Samir. Evolution and the Levels of Selection. Oxford University press, 2006.

Dawkins, R. The Selfish Gene. Oxford University Press, Oxford, 1976, 1989 p. 2.

Tags: additive characters, Alan Turing, causal decomposition, cells, computer science, contextual analysis, cooperation, cross-level by-products, diachronic, emergent characters, evolution, evolutionary game theory, focal level, G. C. Williams, gametes, genes, gene’s-eye perspective, genic selection, George Price, hierarchical organization, kin selection, levels-of-selection controversy, logic, macroevolution, mathematics, maximum deduction, meiotic drive, military code, multi-cellular organisms, multi-level selection 1, multi-level selection 2, pluralism, Price equation, R. A. Fisher, realism, reciprocal altruism, reductionism, Richard Dawkins, Richard Lewontin, selfish genetic element, statistical decomposition, synchronic, W. D. Hamilton

View the original article here

For the first time it right-get selection guide software for small business

Sorry, I was unable fromt the content to this page. Sorry, I was unable fromt the content to this page.

View the original article here

This post was made using the Auto Blogging Software from WebMagnates.org This line will not appear when posts are made after activating the software to full version.

Get it Right the First Time - A Small Business Guide to Software Selection

Introduction
What to Buy - That is the Question
Buying decisions are the essence of life in the commerce-driven 21st century. From everyday decisions like selecting lunch from a restaurant menu, to getting a new car, to major company acquisitions, much of our time is spent "buying".
And these choices are anything but simple. Each marketer professes to be the sole champion of our consumer rights and pummels us with enticing advertising messages, about how their wares are "the best". Seductive as these messages are, no product or service is quite the same. The difference may be glaring - that of "better vs. worse", or a subtle tradeoff between price, quality, feature set, customer service, or durability.
It is therefore important to keep our wits about & develop a systematic approach to the buying decision. Our view should be broad & farsighted, rather than buying based only on what immediately meets the eye. Hasty decisions leave us with flashy features never used, or hefty repair bills of products that came cheap.
A good example of a systematic approach is when you buy a car. A myriad of factors are considered & weighed, which impact the owner for the next decade. This includes brand, performance vs. style, price, safety, terms of finance, mileage, maintenance, resale value & so many other factors.
Selecting Software
In our new "wired" modern reality, software is no less important than products & services in our everyday lives. Whether it's a personal email program, chat software for instant connection, collaboration software to organize scattered employees, or an ERP implementation to manage company processes - there's no surviving without them!
But we're somewhat more used to buying products & services than software, which is a relatively recent phenomenon. In many ways, selecting software is no different from selecting a product or service. Although intangible, software, also address a very real need, on which personal & professional success often depends. Naturally, some of the same purchase factors apply - brand, service, & maintenance costs.
In spite of the patronizing obviousness of the above, software selection is a grey zone; an underdeveloped arena. This accounts for the high incidence of "shelfware" - software that are bought with grand intentions, but end up on dusty shelves. This is because unlike products & services, it is not so intuitively evident that software have "life cycles" & need to be "maintained", "updated", & "repaired".
Therefore, purchases are made based on what immediately meets the eye - technical features. This mistake is understandable, because technical features are well documented & advertised, & easy for the buyer to use as decision criteria. But with this approach, factors that are just as pertinent, but not so immediately obvious, get left out. Some research & serious thinking is needed to gauge these "hidden" factors.
Key Factors To Consider
1) Company History & Experience
The vendor needs to be sized up before we even go on to consider the software itself. Company background is essential because, unlike traditional companies, software companies are often small, & often beyond national boundaries. Since these companies would likely be handling our sensitive data, we need to do a background check. Some related questions are:
How Long Have They Been Around?
As in most cases, we can reasonably assume that past record is a good indicator of future performance. Important questions are - How long have they been around? How long have they been in the field? If they're offering business collaboration software, have they been in this industry long enough? Even if the software is new, do they have experience developing related software?
What is Their Niche?
Does the company know your niche well enough to know your needs? If you are a small/mid sized business, a company mainly serving the Fortune 500 is not for you. If you work from home, it is unlikely a solution serving large offices will meet your needs.
The Ultimate Testament - The Customer
The ultimate judge of software is its users. To get a true picture, it is important to look at how customers are using the software & what their comments are. Does their site include a client's list or page? Check out what customers say under testimonials, or you could even get in touch with the customers yourself for comments.
Dangers
There are certain things about the software industry that a buyer should be wary of. Software startups have shorter life spans than traditional companies & ride high on a success wave, but go "pop" when the industry bubble bursts. This was exemplified by the "dot com burst" of 2000. Whether the current spate of "Web 2.0" companies constitutes another expanding bubble which will inevitably burst is debatable, but it makes sense to be wary & bet your money on dependable companies with proven track records.
2) Cost
There's no denying the importance of cost effectiveness in buying decisions across the board. Yet costs should be seen in a broad perspective, because low entry costs may well result in higher total costs along the product's life.
Features vs. Price
A cost-benefit analysis makes sense, & costs need to be compared with the software's range of features & functionalities. A document management system may not be the cheapest, but it may allow you to also set up a virtual office. Going for loads of features also constitutes a trap, because users never get around to using half of them.
Needs vs. Price
Another question is whether there is an overlap between features & needs at all. Many features may not relate to needs sought to be addressed. You should clearly define your needs, & classify features as "needed features" & "features not needed". Another possible scheme of classifying features could be "must have", "nice to have", & "future requirements".
3) Ease of Use/Adoption
An adoption & learning curve is involved with every new software purchase. It needs to be integrated with current systems & software, & the end users have to be brought up to speed using it. If the software is chunky & too complex, adoption resistance can occur.
Ease of Use
The software should have an intuitive interface, & use of features should be pretty much self evident. The shorter the learning curve training a new user, the better. The software should also have the ability to easily fit into the existing systems with which it will have to communicate. For example, a collaboration software might allow you to use some features from your Outlook itself or even share Outlook data.
Adoption
To get a measure of "shelfware", i.e., software that is purchased but never used, some studies peg the number of shelved content management solutions at 20-25%. At a million dollars per implementation, that's pretty expensive shelfware! According to another study in the US, 22% of purchased enterprise portal (ERP) licenses are never used.
No doubt, "Shelfware" is a result of ill thought out purchase decisions. These studies clearly underline the importance of making an educated purchase. One possible way to protect against shelfware is the new concept of software as a service (SAAS) hosted software. The software is hosted by its developer, & buyers have to pay a monthly subscription, which they can opt out of anytime.
Support
No matter how good a software is, there are bound to be times when one can't find out how to work a particular feature or a glitch crops up. Some software solutions may require you to hire dedicated support staff of your own, while others may be easy to use, and no specialized staff may be needed, and still others may offer free support. The cost of hiring support staff needs to be factored into the buying decision.
Provider support may be in the form of live human support, or automated help engines. In case of human help, the quality of solutions, availability & conduct of support executives matter. Support can also be in the form of an extensively documented help engine, or extensive help information on the company site. This form of support is often more prompt & efficient than human help.
Training
Training is another form of support which deserves special mention. Free training seminars or their new avatar - webinars (online seminars) - greatly help in getting up to speed with the software at no extra cost. In some cases the company might offer paid training, which may be essential, & hence this cost needs to be factored into the purchase decision.
Maintenance
Maintenance costs & efforts have a major impact on the performance & adoptability of software, & hence form important criteria of the buying decision. In case the software is hosted at the company's end, it is of utmost importance that the software be available online at all times, or the "uptime". Uptimes are covered under the "service level agreement" & range from 98% to 99.99%. A minimum uptime of 99% is what one must look for.
The company's upkeep is also important. Efforts to constantly improve upon the software underline a commitment to providing quality service. Are bugs fixed promptly & on an ongoing basis? Are they just releasing software & not updating it? One should develop a habit of keeping up with the company newsletter, release notes or the "what's new" section on their site. Periodic newsletters & a "what's new" section are indicative of a dynamic company.
4) Familiarity
The "feel" of the software is another important criterion. The software should keep with the basic layout & navigation schemes we are used to. This makes for quicker transition.
One good way is to compare with the OS in which we would use the software. Does it have the same basic schema as the OS environment? A software with Mac schema on Windows wouldn't sit that well. Or we could compare it with other software which we are used to. If you are switching to a low priced solution from an expensive one, choosing software with a similar "feel" would make sense. Does it retain most of the main features you are used to?
5) Security
Security is a top consideration because he software company will likely be handling information critical to us - business, financial or personal. We need to be well assured of our data's security & there are no risks of it being compromised. This needs research, & the extensiveness of which depends on the sensitivity of our data.
What safety features does the provider have?
Encryption, or coding of information, is used by most companies to protect the integrity of their clients' information. There are different types of encryption, each of which is associated with a different level of security. DAS is one, once popular but now known to have loopholes. SSL 128-bit encryption is associated with top notch security. Password protection is another important facet. Is the software equipped to withstand manual & automated attempts to hack your password? The ability of the system to detect a hacking attempt & lock up in time is important.
Data Backup
In extreme cases of system breakdown caused by a facility fire, natural disaster or technical glitch etc, it is important that your data is frequently & adequately backed up. Data backup should be frequent & adequate.
Certain factors are to be considered in backup practices. The first is the frequency of backups. If there is a long gap, there is a possibility of data being lost in intermittent periods. Secondly, what are the security arrangements at the facilities where your data resides? Is it manned & guarded by security personnel? What other safeguards are in place? Is there a good firewall? What is the protection against virus attacks? What procedures are in place for disaster management?
Track Record
As with company background, a little research on the security track record makes sense. Has the company ever been vulnerable to attacks before? What were the losses? How did the company react? How many years has the company had a good record? New companies will have a clean record, but that isn't necessarily indicative of good security.
The Server System
The server system where the sensitive data actually lies is very important. Is it state-of-the-art? The server infrastructure could be owned by the software provider themselves or outsourced to a dedicated company providing hosting solutions. Outsourced hosting is a good thing because hosting companies have extensive expertise & infrastructure for security, & this frees up the software provider to concentrate on the software itself. The company might not have an elaborate setup at all, running the software & processing data through computers set up in the garage somewhere acting as servers. This should get your alarm bells ringing!
Conclusion - A Systematic Selection Approach
Now that we have discussed all the relevant factors in detail & have a better perspective of the subject, it is important to develop a systematic approach to analyzing these factors.
What factors are important to me?
Although all of the above factors are relevant, their relative importance may differ from customer to customer. For a company with deep pockets, price comes lower in the list. For a company using collaboration software to process business information, security is high priority. Again if a solution forms an important part of a company's business, it is important that it integrates well with existing systems. For dynamic industries like real estate, short training times are important.
Know Thy Software
By this step you would have selected software. But that is still not the end. For all our theorizing & researching, the software still has to pass its toughest test. Most software allows you a free trial period. It would be a good idea to seriously use this period to analyze the software.
It is important to stay focused during this testing period because the impact is going to be long lasting. Follow systematic planning. Identify objectives & needs, develop a testing plan, lay out the timelines and designate people from different departments to try out different features. Set responsibilities & goals so that testers take their job seriously.
THE DECISION!
Don't hesitate to put the burden onto the company to prove itself. Let the company prove to you the features that seem important to you. For example, if security is of prime importance, ask the company to display how their solution scores high on security. Don't hesitate to call them if you have questions.
Test their service levels to see if it lives up to their promises. If you submit a ticket, is it promptly responded to? Is a good solution provided? If the problem requires live help, do you get it fast enough? When you call in with a problem, is it a live person or an automated message you converse with?
This is as extensively as you can analyze software. You're educated enough to make a choice which will most likely not fail you. You shall surely not be disappointed in your decision.

Get it Right the First Time - A Small Business Guide to Software Selection

Introduction
What to Buy - That is the Question
Buying decisions are the essence of life in the commerce-driven 21st century. From everyday decisions like selecting lunch from a restaurant menu, to getting a new car, to major company acquisitions, much of our time is spent "buying".
And these choices are anything but simple. Each marketer professes to be the sole champion of our consumer rights and pummels us with enticing advertising messages, about how their wares are "the best". Seductive as these messages are, no product or service is quite the same. The difference may be glaring - that of "better vs. worse", or a subtle tradeoff between price, quality, feature set, customer service, or durability.
It is therefore important to keep our wits about & develop a systematic approach to the buying decision. Our view should be broad & farsighted, rather than buying based only on what immediately meets the eye. Hasty decisions leave us with flashy features never used, or hefty repair bills of products that came cheap.
A good example of a systematic approach is when you buy a car. A myriad of factors are considered & weighed, which impact the owner for the next decade. This includes brand, performance vs. style, price, safety, terms of finance, mileage, maintenance, resale value & so many other factors.
Selecting Software
In our new "wired" modern reality, software is no less important than products & services in our everyday lives. Whether it's a personal email program, chat software for instant connection, collaboration software to organize scattered employees, or an ERP implementation to manage company processes - there's no surviving without them!
But we're somewhat more used to buying products & services than software, which is a relatively recent phenomenon. In many ways, selecting software is no different from selecting a product or service. Although intangible, software, also address a very real need, on which personal & professional success often depends. Naturally, some of the same purchase factors apply - brand, service, & maintenance costs.
In spite of the patronizing obviousness of the above, software selection is a grey zone; an underdeveloped arena. This accounts for the high incidence of "shelfware" - software that are bought with grand intentions, but end up on dusty shelves. This is because unlike products & services, it is not so intuitively evident that software have "life cycles" & need to be "maintained", "updated", & "repaired".
Therefore, purchases are made based on what immediately meets the eye - technical features. This mistake is understandable, because technical features are well documented & advertised, & easy for the buyer to use as decision criteria. But with this approach, factors that are just as pertinent, but not so immediately obvious, get left out. Some research & serious thinking is needed to gauge these "hidden" factors.
Key Factors To Consider
1) Company History & Experience
The vendor needs to be sized up before we even go on to consider the software itself. Company background is essential because, unlike traditional companies, software companies are often small, & often beyond national boundaries. Since these companies would likely be handling our sensitive data, we need to do a background check. Some related questions are:
How Long Have They Been Around?
As in most cases, we can reasonably assume that past record is a good indicator of future performance. Important questions are - How long have they been around? How long have they been in the field? If they're offering business collaboration software, have they been in this industry long enough? Even if the software is new, do they have experience developing related software?
What is Their Niche?
Does the company know your niche well enough to know your needs? If you are a small/mid sized business, a company mainly serving the Fortune 500 is not for you. If you work from home, it is unlikely a solution serving large offices will meet your needs.
The Ultimate Testament - The Customer
The ultimate judge of software is its users. To get a true picture, it is important to look at how customers are using the software & what their comments are. Does their site include a client's list or page? Check out what customers say under testimonials, or you could even get in touch with the customers yourself for comments.
Dangers
There are certain things about the software industry that a buyer should be wary of. Software startups have shorter life spans than traditional companies & ride high on a success wave, but go "pop" when the industry bubble bursts. This was exemplified by the "dot com burst" of 2000. Whether the current spate of "Web 2.0" companies constitutes another expanding bubble which will inevitably burst is debatable, but it makes sense to be wary & bet your money on dependable companies with proven track records.
2) Cost
There's no denying the importance of cost effectiveness in buying decisions across the board. Yet costs should be seen in a broad perspective, because low entry costs may well result in higher total costs along the product's life.
Features vs. Price
A cost-benefit analysis makes sense, & costs need to be compared with the software's range of features & functionalities. A document management system may not be the cheapest, but it may allow you to also set up a virtual office. Going for loads of features also constitutes a trap, because users never get around to using half of them.
Needs vs. Price
Another question is whether there is an overlap between features & needs at all. Many features may not relate to needs sought to be addressed. You should clearly define your needs, & classify features as "needed features" & "features not needed". Another possible scheme of classifying features could be "must have", "nice to have", & "future requirements".
3) Ease of Use/Adoption
An adoption & learning curve is involved with every new software purchase. It needs to be integrated with current systems & software, & the end users have to be brought up to speed using it. If the software is chunky & too complex, adoption resistance can occur.
Ease of Use
The software should have an intuitive interface, & use of features should be pretty much self evident. The shorter the learning curve training a new user, the better. The software should also have the ability to easily fit into the existing systems with which it will have to communicate. For example, a collaboration software might allow you to use some features from your Outlook itself or even share Outlook data.
Adoption
To get a measure of "shelfware", i.e., software that is purchased but never used, some studies peg the number of shelved content management solutions at 20-25%. At a million dollars per implementation, that's pretty expensive shelfware! According to another study in the US, 22% of purchased enterprise portal (ERP) licenses are never used.
No doubt, "Shelfware" is a result of ill thought out purchase decisions. These studies clearly underline the importance of making an educated purchase. One possible way to protect against shelfware is the new concept of software as a service (SAAS) hosted software. The software is hosted by its developer, & buyers have to pay a monthly subscription, which they can opt out of anytime.
Support
No matter how good a software is, there are bound to be times when one can't find out how to work a particular feature or a glitch crops up. Some software solutions may require you to hire dedicated support staff of your own, while others may be easy to use, and no specialized staff may be needed, and still others may offer free support. The cost of hiring support staff needs to be factored into the buying decision.
Provider support may be in the form of live human support, or automated help engines. In case of human help, the quality of solutions, availability & conduct of support executives matter. Support can also be in the form of an extensively documented help engine, or extensive help information on the company site. This form of support is often more prompt & efficient than human help.
Training
Training is another form of support which deserves special mention. Free training seminars or their new avatar - webinars (online seminars) - greatly help in getting up to speed with the software at no extra cost. In some cases the company might offer paid training, which may be essential, & hence this cost needs to be factored into the purchase decision.
Maintenance
Maintenance costs & efforts have a major impact on the performance & adoptability of software, & hence form important criteria of the buying decision. In case the software is hosted at the company's end, it is of utmost importance that the software be available online at all times, or the "uptime". Uptimes are covered under the "service level agreement" & range from 98% to 99.99%. A minimum uptime of 99% is what one must look for.
The company's upkeep is also important. Efforts to constantly improve upon the software underline a commitment to providing quality service. Are bugs fixed promptly & on an ongoing basis? Are they just releasing software & not updating it? One should develop a habit of keeping up with the company newsletter, release notes or the "what's new" section on their site. Periodic newsletters & a "what's new" section are indicative of a dynamic company.
4) Familiarity
The "feel" of the software is another important criterion. The software should keep with the basic layout & navigation schemes we are used to. This makes for quicker transition.
One good way is to compare with the OS in which we would use the software. Does it have the same basic schema as the OS environment? A software with Mac schema on Windows wouldn't sit that well. Or we could compare it with other software which we are used to. If you are switching to a low priced solution from an expensive one, choosing software with a similar "feel" would make sense. Does it retain most of the main features you are used to?
5) Security
Security is a top consideration because he software company will likely be handling information critical to us - business, financial or personal. We need to be well assured of our data's security & there are no risks of it being compromised. This needs research, & the extensiveness of which depends on the sensitivity of our data.
What safety features does the provider have?
Encryption, or coding of information, is used by most companies to protect the integrity of their clients' information. There are different types of encryption, each of which is associated with a different level of security. DAS is one, once popular but now known to have loopholes. SSL 128-bit encryption is associated with top notch security. Password protection is another important facet. Is the software equipped to withstand manual & automated attempts to hack your password? The ability of the system to detect a hacking attempt & lock up in time is important.
Data Backup
In extreme cases of system breakdown caused by a facility fire, natural disaster or technical glitch etc, it is important that your data is frequently & adequately backed up. Data backup should be frequent & adequate.
Certain factors are to be considered in backup practices. The first is the frequency of backups. If there is a long gap, there is a possibility of data being lost in intermittent periods. Secondly, what are the security arrangements at the facilities where your data resides? Is it manned & guarded by security personnel? What other safeguards are in place? Is there a good firewall? What is the protection against virus attacks? What procedures are in place for disaster management?
Track Record
As with company background, a little research on the security track record makes sense. Has the company ever been vulnerable to attacks before? What were the losses? How did the company react? How many years has the company had a good record? New companies will have a clean record, but that isn't necessarily indicative of good security.
The Server System
The server system where the sensitive data actually lies is very important. Is it state-of-the-art? The server infrastructure could be owned by the software provider themselves or outsourced to a dedicated company providing hosting solutions. Outsourced hosting is a good thing because hosting companies have extensive expertise & infrastructure for security, & this frees up the software provider to concentrate on the software itself. The company might not have an elaborate setup at all, running the software & processing data through computers set up in the garage somewhere acting as servers. This should get your alarm bells ringing!
Conclusion - A Systematic Selection Approach
Now that we have discussed all the relevant factors in detail & have a better perspective of the subject, it is important to develop a systematic approach to analyzing these factors.
What factors are important to me?
Although all of the above factors are relevant, their relative importance may differ from customer to customer. For a company with deep pockets, price comes lower in the list. For a company using collaboration software to process business information, security is high priority. Again if a solution forms an important part of a company's business, it is important that it integrates well with existing systems. For dynamic industries like real estate, short training times are important.
Know Thy Software
By this step you would have selected software. But that is still not the end. For all our theorizing & researching, the software still has to pass its toughest test. Most software allows you a free trial period. It would be a good idea to seriously use this period to analyze the software.
It is important to stay focused during this testing period because the impact is going to be long lasting. Follow systematic planning. Identify objectives & needs, develop a testing plan, lay out the timelines and designate people from different departments to try out different features. Set responsibilities & goals so that testers take their job seriously.
THE DECISION!
Don't hesitate to put the burden onto the company to prove itself. Let the company prove to you the features that seem important to you. For example, if security is of prime importance, ask the company to display how their solution scores high on security. Don't hesitate to call them if you have questions.
Test their service levels to see if it lives up to their promises. If you submit a ticket, is it promptly responded to? Is a good solution provided? If the problem requires live help, do you get it fast enough? When you call in with a problem, is it a live person or an automated message you converse with?
This is as extensively as you can analyze software. You're educated enough to make a choice which will most likely not fail you. You shall surely not be disappointed in your decision.