$0 #5 "Number of reactants" #a "Matrix of stoichiometric coefficients" #f "Types of calibration functions" #10 "Values of wavelengths, wave numbers or frequencies" #14 "Values of attributes of linear {calibration} parameters" #15 'Definition of a calibration function of type -6 (enumeration of active reactants)' #19 "Number of solutions" #1e "Values of attributes of nonlinear {chemical} parameters" 'Values of attributes of optionally individual parameters' #1a "Number of vessels" #1b "Number of burettes" #1f "Values of attributes of nonlinear {chemical} parameters" 'Values of attributes of definitely common parameters' #23 "Initial estimates of unknown nonlinear {chemical} parameters" #24 "Values of known definitely common parameters" #28 "Request for services" #2a "Common input data of services" 'Calculation of the ionic strength of a solution' #2b "Common input data of services" 'Ridge estimation of parameters in a linear model' #2c "Common input data of services" 'Setting a non-standard significance level for statistical tests' #2d "Common input data of services" 'Significance test of estimate of parameters' #32 "Number of a calibration function" #37 "Values of specifiers of a calibration function" #3c "Values of known linear {calibration} parameters" #41 "Number of a solution" #46 "Values of known optionally individual nonlinear {chemical} parameters" #42 "Number of a vessel" #43 "Number of a burette" #4b "Individual input data of services" 'Calculation of the ionic strength of a solution' #50 "Estimates of the standard deviations of the quantities on abscissa and ordinate" #51 "Estimate of the standard deviation of the quantity on ordinate" #55 "Measured values of the quantities on abscissa and ordinate" #56 "Measured values of the quantity on abscissa" #57 "Measured values of the quantity on ordinate" #5a 'Number of the section actual section is linked to' #5c 'Number of principal components of an "absorbance" matrix' $1 #9 =0! #2d 0! #3f =0! #80 %Recognized format of the input file doesn't correspond to the format of the expected item (only formal error occurs at the marked position, logical error can be found anywhere "above"). $2 #1 >Number of reactants can't be a negative number! #2 >Number of reactants can't be zero! #3 >Multiplier can't be a negative number! #4 >Multiplier can't be zero! #8 >It is not possible to "multiply" a calibration function of type 9, -1, -2 or -6 (by the way, it wouldn't make a sense)! #9 >In definition of an active component (we are talking about a calibration function of type -6) it is not possible to include a reactant twice! #a >There's not any predefined type of a calibration function related to the given number! Permitted values lie in the closed intervals <1,9> and <-1,-7>, three predefined logarithmic subtypes can be selected by means of the 'ln' (or simply 'l') postfix, i.e. 9ln, -5ln, -6ln (or 9l, -5l, -6l). #b >There's not any reactant related to the given number! #c >There's not any species related to the given number! #d >There's not any component related to the given number! #e >Number defining how many active components (we are talking about a calibration function of type -5 or -6) a reactant contains must be natural! #f >Number defining how many active components (we are talking about a calibration function of type -5) a reactant contains must be assigned the same value every time it is specified! We can type e.g. ?2=-2 3=-2; ?2=-2\1 3=-2; ?2=-2\3 3=-2\3; (\1 needn't be specified, it holds implicitly) but never e.g. ?2=-2 3=-2\2; ?2=-2\3 3=-2\2; #10 >It is not possible to mark a parameter of a calibration function of type 7, 8 or -5 unknown (defined), if it is (has been) defined (by means of unknown parameters and/or given real numbers). #11 >It is not possible to mark a parameter of a calibration function of type 7, 8 or -5 defined (by means of unknown parameters and/or given real numbers), if it is unknown. #14 >It is not possible to mark a parameter of a calibration function of type 7, 8 or -5 defined (by means of unknown parameters and/or given real numbers), if it is defining. #17 >It is not possible to mark a parameter of a calibration function of type 7, 8 or -5 defining, if it is defined (by means of unknown parameters and/or given real numbers). #1a >It is not possible to mark a parameter of a calibration function of type 7, 8 or -5 known, if it is unknown (defining) or defined (by means of unknown parameters and/or given real numbers). #1b >Attribute of a linear parameter must have a value equal either to 0 (if the parameter is known), 1 (if the parameter is unknown) or -1 (if the parameter is known and zero). #1c >Number of solutions can't be a negative number! #70 >Number of vessels can't be a negative number! #71 >Number of burettes can't be a negative number! #1d >Number of solutions can't be zero! #72 >Number of vessels can't be zero! #73 >Number of burettes can't be zero! #1e >It is not possible to mark a chemical parameter unknown (defined), if it is (has been) defined (by means of unknown parameters and/or given real numbers). #1f >It is not possible to mark a chemical parameter defined (by means of unknown parameters and/or given real numbers), if it is unknown. #20 >It is not possible to mark a chemical parameter defined (by means of unknown parameters and/or given real numbers), if it is defining. #21 >It is not possible to mark a chemical parameter defining, if it is defined (by means of unknown parameters and/or given real numbers). #22 >It is not possible to mark a chemical parameter known and zero, if it is unknown (defining) or defined (by means of unknown parameters and/or given real numbers). #24 >Attribute of a nonlinear optionally individual parameter must have a value equal either to 0 (if the parameter is known), 1 (if the parameter is unknown) or -1 (if the parameter is known and zero). #25 >Attribute of a nonlinear definitely common parameter must have a value equal either to 0 (if the parameter is known) or 1 (or 2) (if the parameter is unknown). *Attribute with a value equal to 2 marks an unknown stability constant whose estimate will be calculated (if possible) in logarithmic scale. #26 >Stability constant can't be a negative number! #27 >Volume can't be a negative number! #28 >Mass amount of a reactant can't be a negative number, if there's not any negative element (or an unknown element with negative initial estimate) in the related column of the matrix of stoichiometric coefficients! #29 >Mass amount of a reactant can't be a negative number, if there's not any species defined! #2a >Analytical concentration of a reactant can't be a negative number, if there's not any negative element (or an unknown element with negative initial estimate) in the related column of the matrix of stoichiometric coefficients! #2b >Analytical concentration of a reactant can't be a negative number, if there's not any species defined! #2d >Measure of bias of ridge estimate in a linear model can't be a negative number! #2e >Significance level must lie in the open interval (0,1)! #2f >There's not any unknown parameter related to the given number! #32 >It is not possible to select a parameter twice! #35 >Number of "electrons" in the definition of a Nernstian slope can't be zero! #38 >There's not any calibration function related to the given number (the functions are numbered from 1)! #3a >It is not possible to select a reactant twice! #3b >It is not possible to select a species twice! #3c >It is not possible to select a component twice! #3d >"Thickness" of a solution (optical length) can't be a negative number! #3e >"Thickness" of a solution (optical length) can't be zero! #41 >There's not any solution related to the given number (the solutions are numbered from 1)! #42 >It is not possible to "leave out" the analytical concentration of a reactant in a solution of the second type (negative logarithm of the free concentration of the reactant in the solution must be given directly at this position in the form e.g. =1.736)! #43 >It is not possible to "leave out" the analytical concentration of a reactant in a solution, if the analytical concentration of the reactant in the burette is unknown! #74 >There's not any vessel related to the given number (the vessels are numbered from 1)! #75 >There's not any burette related to the given number (the burettes are numbered from 1)! #44 >Double the ionic strength of inert ions present in a solution can't be a negative number! #45 >Double the ionic strength of inert ions present in a burette can't be a negative number! #47 >Estimate of the standard deviation of the quantity on abscissa can't be a negative number! #48 >Estimate of the standard deviation of the quantity on ordinate (set absolutely) can't be a negative number! #49 >Estimates of the standard deviations of the quantities on abscissa and ordinate can't be both zero! If you want to "cancel" the variance-covariance matrix of the dependent variable, set 0 and 1 (in this order). #4a >Estimate of the standard deviation of the quantity on ordinate (set absolutely) can't be zero, if a solution of the second type is assigned to the actual section (the burette is not defined)! If you want to "cancel" the variance-covariance matrix of the dependent variable, set 1. #4b >Estimate of the standard deviation of the quantity on ordinate (set relatively in per cents of a measured value of the quantity) must lie in the interval <0,100)! #4d >There's not any section related to the given number (the sections are numbered from 1)! #4e >There's not any section related to the given number (it is not possible to select a section which hasn't been defined yet)! #4f >It is not possible to copy individual data from the previous section (by the ^, " or ~ control character), if the actual section is "linked" and if it is linked neither to the previous section nor to the section the previous section is linked to (if the previous section is "linked", too)! #50 >It is not possible to copy an estimate of the standard deviation of the quantity on ordinate (set either absolutely or relatively in per cents of a measured value of the quantity) from the previous section (by the " or ' control character or (if a solution of the second type is assigned to the actual section) implicitly by leaving out the & control character and repeating the item "Measured values of the quantity on ordinate" or by means of a template of analytical and/or free concentrations), if the calibration functions assigned to the actual and the previous sections are not of "similar" types! *Groups of "similar" types are {1, 2, 3, 4, 5, 6}, {7, 8}, {9}, {9ln}, {-1, -2}, {-3, -4}, {-5}, {-5ln}, {-6}, {-6ln}, {-7}. Moreover, if all the parameters of a calibration function of type -5 are either known and zero (with attribute -1) or defined by a constant boundary condition with a value equal either to 0 or 1, the function is "similar" to a calibration function of type -6. #51 >It is not possible to copy an estimate of the standard deviation of the quantity on ordinate (set both absolutely and relatively in per cents of a measured value of the quantity) from the previous section (by the " or ' control character or (if a solution of the second type is assigned to the actual section) implicitly by leaving out the & control character and repeating the item "Measured values of the quantity on ordinate" or by means of a template of analytical and/or free concentrations), if calibration functions of type -6 (or -5 and -6) are assigned to the actual and the previous sections while one of them (of type -6) transforms the measured values of the quantity on ordinate from the interval <0,100> to the interval <0,1> (e.g. the % control character is set preceding the value of the only specifier of the function)! #52 >It is not possible to set a number of a calibration function by the , (comma) control character, if the function with the highest possible number (defined implicitly in the item "Types of calibration functions") has been assigned to the previous section! Comma "adds" one to the number of the calibration function assigned to the previous section sometimes putting us aside the defined set of functions. #53 >It is not possible to set a number of a calibration function implicitly by means of a template of added volume (starting with the " or ~ control character) and a new row of an "absorbance" matrix, if the function with the highest possible number (defined implicitly in the item "Types of calibration functions") has been assigned to the previous section! Quotation marks or tilde ensure increasing of the number of a calibration function by one every time the template is 'filled up' (a new row of an "absorbance" matrix is reached) sometimes putting us aside the defined set of functions. #54 >It is not possible to set a number of a calibration function implicitly by means of a template of analytical and/or free concentrations and a new row of an "absorbance" matrix, if the function with the highest possible number (defined implicitly in the item "Types of calibration functions") has been assigned to the previous section (point)! The use of a template of analytical and/or free concentrations ensures increasing of the number of a calibration function by one every time the template is 'filled up' (a new row of an "absorbance" matrix is reached) sometimes putting us aside the defined set of functions. #55 >It is not possible to set a number of a calibration function (if a solution of the second type is assigned to the actual section) implicitly by leaving out the & control character and repeating the item "Measured values of the quantity on ordinate", if the function with the highest possible number (defined implicitly in the item "Types of calibration functions") has been assigned to the previous section (point)! Leaving out the & control character ensures increasing of the number of a calibration function every time a new point is reached sometimes putting us aside the defined set of functions. #56 >It is not possible to set a number of a solution by the , (comma) control character, if the solution with the highest possible number (defined implicitly in the item "Number of solutions") is assigned to the previous section! Comma "adds" one to the number of the solution assigned to the previous section sometimes putting us aside the defined set of solutions. #76 >It is not possible to set a number of a vessel by the , (comma) control character, if the vessel with the highest possible number (defined implicitly in the item "Number of vessels") is assigned to the previous section! Comma "adds" one to the number of the vessel assigned to the previous section sometimes putting us aside the defined set of vessels. #77 >It is not possible to set a number of a burette by the , (comma) control character, if the burette with the highest possible number (defined implicitly in the item "Number of burettes") is assigned to the previous section! Comma "adds" one to the number of the burette assigned to the previous section sometimes putting us aside the defined set of burettes. #57 >Measured value of the quantity on ordinate can't be a negative number, if a calibration function of type -6 (or -6ln) is assigned to the actual section! #58 >Measured value of the quantity on ordinate can't be greater than 1, if a calibration function of type -6 is assigned to the actual section! If you want to give the values in %, use the possibility to transform the measured values of the quantity on ordinate from the interval <0,100> to the interval <0,1> (set the % control character to precede the value of the only specifier of the function). #59 >Measured value of the quantity on ordinate can't be greater than 100, if a calibration function of type -6 (transforming the measured values of the quantity on ordinate from the interval <0,100> to the interval <0,1>) is assigned to the actual section! #5a >Added volume must grow in every section. Better control over contingent errors is gained when defining "linked" sections or sections with zero free concentrations of some reactants in the first point. #5b >If there's a zero free concentration of any reactant in some point of a section (according to some other limitations only the first point comes into consideration), the free concentration must be zero in all the other points of the section, too. If it is not true, it is necessary to make a separate section for the first point. In the simplest case it can be gained by putting the & ^ or @ ^ control characters between the first and the second points of the section, but generally it can be more difficult (e.g. with "linked" or "copying" sections the relations between the sections must be redefined). #5c >Section from which the measured values of the quantity on abscissa are copied doesn't contain more points. Not further than at this position the actual section must be completed, too (what more points would you like to copy?). #5d >Section from which the measured values of the quantity on abscissa are copied doesn't contain more points. Not further than at this position the actual section must be completed, too (there are no more points to leave out!). #5e >It is not possible to set a value of a specifier of a calibration function by the ' (apostrophe) control character, if the types of the functions assigned to the actual and the previous sections do not coincide (apostrophe copies the value from the previous section)! #5f >It is not possible to set a value of a parameter of a calibration function by the ' (apostrophe) control character, if one and the same function is not assigned to both the actual and the previous section (apostrophe copies the value from the previous section)! #60 >It is not possible to set a value of a parameter of a calibration function by the ' (apostrophe) control character, if one and the same function is not assigned to both the actual and the previous section (a coincidence of types of two different functions doesn't suffice)! #61 >It is not possible to set the analytical concentration (mass amount) of a reactant in a solution by the ' (apostrophe) control character, if one and the same solution is not assigned to both the actual and the previous section! #78 >It is not possible to set the volume of a solution in a vessel by the ' (apostrophe) control character, if one and the same vessel is not assigned to both the actual and the previous section! #79 >It is not possible to set the analytical concentration of a reactant in a burette by the ' (apostrophe) control character, if one and the same burette is not assigned to both the actual and the previous section! #62 >It is not possible to set the analytical concentration of a reactant in a solution by the ' (apostrophe) control character, if the corresponding analytical concentration is "left out" in the previous section! #63 >Number of principal components can't be a negative number! #64 >Number of principal components can't be zero! #65 >Number of principal components can't be greater than number of columns of the "absorbance" matrix! #66 >Number of principal components can't be greater than number of rows of the "absorbance" matrix! $3 #7 >It is not possible to define a calibration function of type 1, 3 or 5, if there's not any species defined! #a >There's no other way how to define a parameter of a calibration function of type -5ln than by a constant boundary condition with zero value (at most one of the parameters of a function of type -5ln can be different from zero)! *Just one of the parameters of a calibration function of type -5ln must be either known (with attribute 0) or unknown (with attribute 1), all the other parameters must be either known and zero (with attribute -1) or defined by a constant boundary condition with zero value. #e >At least one (mostly just one) reactant must be defined for every calibration function of type -5 as a part of an active component. It can be done implicitly by means of attributes of parameters of the calibration function. It holds, that among all attributes of parameters corresponding to individual reactants (not species!) at least one must have a value different from -1, i.e. 0 (if the parameter is known) or 1 (if the parameter is unknown). It is possible to set (by the \ control character and a natural number, both following closely a value of the attribute) how many active components a reactant contains (implicitly, i.e. if the \ control character is not used, just one). #f >Species which doesn't contain an active reactant can't be active itself, i.e. a parameter of a calibration function of type -5 with attribute other than -1 mustn't correspond to it (a parameter of a calibration function of type -5 with attribute -1 must correspond to it)! #12 >At most one of the parameters of a calibration function of type -5ln can be different from zero! *Just one of the parameters of a calibration function of type -5ln must be either known (with attribute 0) or unknown (with attribute 1), all the other parameters must be either known and zero (with attribute -1) or defined by a constant boundary condition with zero value. #13 *Just one of the parameters of a calibration function of type -5ln must be either known (with attribute 0) or unknown (with attribute 1), all the other parameters must be either known and zero (with attribute -1) or defined by a constant boundary condition with zero value. #15 >It is not possible to define a "binding" between an unknown free concentration and any other parameter! #16 >It is not possible to define a "binding" between an unknown analytical concentration and an unknown mass amount! #1b >Specifier of a calibration function of type -6 (or {if the ? control character is used} -5ln) can't have a value corresponding to a reactant which hasn't been marked active! #1c >Specifier of a calibration function of type -6 (or {if the ? control character is used} -5ln) can't have a value corresponding to a species which doesn't contain any active reactant! #1f >Defined mass amount of a reactant (see the item "Values of attributes of nonlinear {chemical} parameters") can't be a negative number, if there's not any negative element (or an unknown element with negative initial estimate) in the related column of the matrix of stoichiometric coefficients! #20 >Defined analytical concentration of a reactant (see the item "Values of attributes of nonlinear {chemical} parameters") can't be a negative number, if there's not any negative element (or an unknown element with negative initial estimate) in the related column of the matrix of stoichiometric coefficients! #21 >It is not possible to assign a calibration function of type -5 or -6 and a solution of the second type to a section, if the analytical concentration in denominator of the function is unknown! In such a case it is necessary either to resign the request for the calculation of this analytical concentration or to "fix" it inside the function, i.e. to replace the function of type -5 or -6 with function of type 8 or -4 (the only parameter is known and equal to 1) and to substitute the reciprocal value of the best estimate of the unknown analytical concentration for a value of "thickness" of a solution (optical length). Generally, both the functions are equivalent, if the analytical concentration in denominator of the function of type -5 or -6 is constant and equal to the reciprocal value of the specifier of the function of type 8 or -4 (the only parameter is known and equal to 1). #23 >There's not any unknown parameter defined! What do you exactly want to calculate? #2a >Each section must contain at least one point! #2b >Section to which a solution of the second type is assigned must contain just (at least) one point! #2c >Template of added volume must be (partially) 'filled up' at least once! #2d >Template of analytical and/or free concentrations must be 'filled up' at least once! #30 >It is not possible to make a link (to any section) if there's not any solution of the first type defined! #31 >It is not possible to make a link to a section to which a solution of the second type is assigned! #38 >If a known and zero volume of a solution is defined in a section, then the added volume can't be zero in any point of the section (according to some other limitations only the first point comes into consideration). #39 >If an unknown volume of a solution in a vessel is defined with zero initial estimate, then the added volume can't be zero in any point (according to some other limitations only the first point comes into consideration) of any section the vessel is assigned to. #3b >Template of analytical and/or free concentrations mustn't include a solution of the first type! #45 >Number of points must be greater than number of unknown parameters! #46 >At least one of the calibration functions is not assigned to any section. All "redundant" definitions must be removed! #47 >Number of unknown linear {calibration} parameters of each function must be less than total number of points in sections the function is assigned to! #48 >At least one of the solutions is not assigned to any section. All "redundant" definitions must be removed! #49 >Number of unknown parameters of each solution must bee less than total number of points in sections the solution is assigned to! #70 >At least one of the vesselss is not assigned to any section. All "redundant" definitions must be removed! #71 >Number of unknown parameters of each vessel must bee less than total number of points in sections the vessel is assigned to! #72 >At least one of the burettes is not assigned to any section. All "redundant" definitions must be removed! #73 >Number of unknown parameters of each burette must bee less than total number of points in sections the burette is assigned to! #4a >At least one of the "known" species is not present in any section. All "redundant" definitions must be removed! #4b >At least one of the "unknown" species is not present in any section. All "redundant" definitions must be removed! #4c >Each "unknown" species must be present in at least two points! #4d >At least one of the "unknown" species is not present in a sufficient number of points! #52 >It is not possible to request the principal component analysis of an "absorbance" matrix, if a variable variance-covariance matrix of the dependent variable is defined! #53 >It is not possible to request the principal component analysis of an "absorbance" matrix, if the template of added volume doesn't start with the " control character (quotation marks ensure increasing of the number of a calibration function by one every time the template is 'filled up' and {in contradistinction to the ~ control character} also copying an estimate of the standard deviation of the measured quantity on ordinate)! *In the principal component analysis of an "absorbance" matrix suitable linear combinations of all original wavelengths ("active" nuclei) are computed which causes some lost of uniqueness of the defined calibration functions (all involved functions must be of the same type having the same values of the attributes of all their parameters). #54 >Number of principal components of an "absorbance" matrix can't be greater than number of defined calibration functions of type either 7 or -5 arranged in a single continuous block starting with the function assigned to the actual section! *In the principal component analysis of an "absorbance" matrix suitable linear combinations of all original wavelengths ("active" nuclei) are computed which causes some lost of uniqueness of the defined calibration functions (all involved functions must be of the same type having the same values of the attributes of all their parameters, which can be achieved most easily by their "multiplication", e.g. >3*7; etc.). #5b >It is possible to request the principal component analysis only if the attributes of all the parameters of all involved calibration functions are the same! *In the principal component analysis of an "absorbance" matrix suitable linear combinations of all original wavelengths ("active" nuclei) are computed which causes some lost of uniqueness of the defined calibration functions (all involved functions must be of the same type having the same values of the attributes of all their parameters, which can be achieved most easily by their "multiplication", e.g. >3*7; etc.). #60 >It is possible to request the principal component analysis only if the same (or no) additive boundary conditions (moreover, without use of constants, it is not possible to type e.g. 1=2&~68.2 etc.) for all unknown parameters of all involved calibration functions are defined! *In the principal component analysis of an "absorbance" matrix suitable linear combinations of all original wavelengths ("active" nuclei) are computed which causes some lost of uniqueness of the defined calibration functions (all involved functions must be of the same type having the same values of the attributes of all their parameters, which can be achieved most easily by their "multiplication", e.g. >3*7; etc.). #61 >It is not possible to request the principal component analysis of an "absorbance" matrix, if a boundary condition containing a constant (in the item "Values of attributes of linear {calibration} parameters") has been defined for one of the linear parameters (of each involved calibration function)! Additive "bindings" without use of constants (e.g. 1=2&2*3 etc.) are permitted. *In the principal component analysis of an "absorbance" matrix suitable linear combinations of all original wavelengths ("active" nuclei) are computed. Constant boundary conditions (e.g. 1=~34.7 etc.) can be applied using known linear parameters (with attribute 0), variable boundary conditions (e.g. 1=2&~68.2 etc.) can't be applied. $7 #80 >Not enough memory!