Re: Properly formatted large numbers
Thanks Terry and Alex. Yes, I suspect that your suggestions would work, but both involve me working out how to trap clicks in the datagrid. That seems like more effort than I can put in at the moment: I've found datagrids to be a bit of a mystery so far and teaching starts here next week :-( This is a very nice example from today's list digest that shows why this issue can be important: Hi Jacque, I made the suggestion I did because at 100 Million records plus 100 million? Yes, well...I think I read the zeros wrong. Should there not be a numberformat setting that formats the numbers for human readability using the system setting of delimiter? Regards, Michael Dear Listers I'm working on some statistical simulations and regularly get output numbers with anything from 1 to 7 digits. They are hard to read when they don't have the conventional commas separating the thousands and millions. I've written a simple function that does the comma formatting for me but it mucks up sorting. Of course. (I'm using a datagrid to display the results and live sorting is really handy.) Terry Judd responded: Can you have two columns - one formatted, one not - and somehow apply the sort to the unformatted column. I guess the unformatted column would have to be of invisible and you'd need to have some way of trapping the selection of the column header. You might need Trevor's help there. Terry... Alext Tweedy suggested: I've not yet used a datagrid, so this is a guess can you supply a custom sort function ? Something like function testLessThan p1, p2 replace comma with empty in p1 replace comma with empty in p2 return p1 p2 end testLessThan -- Alex. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Properly formatted large numbers
Dear Listers I'm working on some statistical simulations and regularly get output numbers with anything from 1 to 7 digits. They are hard to read when they don't have the conventional commas separating the thousands and millions. I've written a simple function that does the comma formatting for me but it mucks up sorting. Of course. (I'm using a datagrid to display the results and live sorting is really handy.) Does anyone have an easy way to deal with this? I'm tempted to use a fixed pitch font so that larger numbers always look longer, but that would be a kludgey workaround. (I note that in OSX the Finder puts separators in the correct places in large numbers using the settings of the International System Preferences.) Regards, Michael -- Michael J Lew Senior Lecturer Department of Pharmacology email: micha...@unimelb.edu.au phone: +61 3 8344 7812 ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: bad interaction between random and word?
Mick I too have been playing with random numbers (I do lots of statistical programming with Rev, so random(0 is one of my tools). Your problem sound a very much like it relates to a similar issue that I came across with random() and arrays a few days ago (quoted below). I'm pretty sure now that it represents a bug -- a very scary one too, because it would be completely invisible in many circumstances. Do you want to post a bug report? Mike My post: I have struck an unexpected issue with an array. I assumed that the two following code snipets would be equivalent, but they are not. Version 1 put random(10)-1 into tnum add 1 to iHisto[tnum] Version 2 add 1 to iHisto[(random(10)-1)] Version 1 does what I expected, and, I guess is more readable and so preferable for that reason. However, I can't see why the two versions differ in their effects. Here is the full code of my button for anyone who wishes to try it out (the different versions are indicated with comments): on mouseUp pMouseBtnNo put 1 into oreps put 100 into ireps set the cursor to busy repeat oreps put 0 into kount repeat ireps add 1 to kount -- **next two lines -- put random(10)-1 into tnum -- add 1 to iHisto[tnum] -- **OR the next one line add 1 to iHisto[(random(10)-1)] -- end repeat --inner combine iHisto by return and comma sort lines of iHisto numeric descending by item 2 of each add 1 to ohisto[item 2 of line 1 of iHisto] put empty into iHisto end repeat --outer combine oHisto with return and tab sort lines of oHisto numeric ascending by word 1 of each put oHisto into fld OutputFld end mouseUp Thanks, Michael Lew Original post: Hello, everyone, I was working on a project that involved randomness, which seemed to be a little off, so I did some testing. I ran this handler from msgbox: on tst2 repeat with i = 2 to 10 put i into randarg put 0 0 0 0 0 0 0 0 0 0 into theList put 1 * i into nRepeats repeat nRepeats times add 1 to word random(randarg) of theList end repeat put nRepeats randarg theList into line i of fld tmpFld end repeat end tst2 THE RESULTS: 2 2 10027 10027 0 0 0 0 0 0 0 0 3 3 9974 9973 9974 0 0 0 0 0 0 0 4 4 10152 10153 10151 10150 0 0 0 0 0 0 5 5 10144 10145 10142 10143 10142 0 0 0 0 0 6 6 9981 9981 9979 9982 9979 9980 0 0 0 0 7 7 10188 10193 10191 10187 10188 10192 10191 0 0 0 8 8 10088 10089 10088 10089 10087 10086 10088 10087 0 0 9 9 9984 9978 9976 9984 9983 9980 9980 9983 9983 0 10 10 9998 9998 9997 9998 9998 9998 9998 9998 You can see the numbers of repeats are as expected, yet the sum of each line SHOULD be exactly the same as the number of repeats, but isn't (some above, some below) ... the individual numbers are all near to 1 but even nearer to one another in a given line (but aren't all equal either). HOWEVER, when I replace the line add 1 to word random(randarg) of theList WITH put random(randarg) into tmp add 1 to word tmp of theList THE RESULTS ARE: 2 2 9825 10175 0 0 0 0 0 0 0 0 3 3 9792 9994 10214 0 0 0 0 0 0 0 4 4 10077 9955 9808 10160 0 0 0 0 0 0 5 5 9926 10083 10078 9888 10025 0 0 0 0 0 6 6 9964 10100 9922 9928 10019 10067 0 0 0 0 7 7 9920 10094 9945 9823 9968 10079 10171 0 0 0 8 8 10143 10129 10080 9809 10024 9948 9971 9896 0 0 9 9 9764 10054 10094 9983 10038 10017 9991 10070 9989 0 10 10 10029 9947 9910 9776 10252 10114 9935 9938 10006 10093 This appears correct ... the sum of each line is (the ones I've checked are and estimates of the others are close) the number of repeats and the numbers seem randomly sprinkled around 1. It seems to me that the replacement lines should do exactly the same thing as the line they replace. Is this not working correctly or am I missing some knowledge of how it should be working? ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Array issue
Dear listers I have struck an unexpected issue with an array. I assumed that the two following code snipets would be equivalent, but they are not. Version 1 put random(10)-1 into tnum add 1 to iHisto[tnum] Version 2 add 1 to iHisto[(random(10)-1)] Version 1 does what I expected, and, I guess is more readable and so preferable for that reason. However, I can't see why the two versions differ in their effects. Here is the full code of my button for anyone who wishes to try it out (the different versions are indicated with comments): on mouseUp pMouseBtnNo put 1 into oreps put 100 into ireps set the cursor to busy repeat oreps put 0 into kount repeat ireps add 1 to kount -- **next two lines -- put random(10)-1 into tnum -- add 1 to iHisto[tnum] -- **OR the next one line add 1 to iHisto[(random(10)-1)] -- end repeat --inner combine iHisto by return and comma sort lines of iHisto numeric descending by item 2 of each add 1 to ohisto[item 2 of line 1 of iHisto] put empty into iHisto end repeat --outer combine oHisto with return and tab sort lines of oHisto numeric ascending by word 1 of each put oHisto into fld OutputFld end mouseUp Thanks, Michael Lew ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Deleting a column ...
Dear Bob You need to set the columndelimiter to comma for your script to work. The columndelimiter defaults to tab (sensibly, in my opinion). This works: on testFunction breakpoint set the columndelimiter to comma put 1,2,3 return 4,5,6 into myVar split myVar by column delete variable myVar[2] combine myvar by column put myVar end testFunction Michael I did this: ON testFunction breakpoint put 1,2,3 return 4,5,6 into myVar split myVar by column delete variable myVar[2] combine myvar by column put myVar END testFunction ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Gaussian pseudo-random numbers -- math issues
Dear Tim, I have an extensive library of statistical maths routines. This handler and function will get you going with your problem: on mouseUp pMouseBtnNo put fld MeanFld into tmean put fld stdevFld into tstdev put fld sample sizeFld into n put fastNormalDeviates(n) into tdevs split tdevs by return --to make it into an array multiply tdevs by tstdev --to make the scale the stdev to the desired value add tmean to tdevs --to obtain the desired mean combine tdevs with return put tdevs into fld outputFld end mouseUp function fastNormalDeviates howmany --returns random deviates from a normal distribution mean 0, stdev 1. --Fast algorithm simply averages many deviates from uniform distributions. --Many times faster than Gasdev-based algorithms. --Values returned are return-delimited. if howmany is empty then put 1 into howmany repeat howmany repeat 20 add random(1) to tsum end repeat put (5000.5-(tsum/20))/645 return after tvals put empty into tsum end repeat delete last char of tvals return tvals end fastNormalDeviates Regards, Michael -- Michael J Lew Senior Lecturer Department of Pharmacology email: [EMAIL PROTECTED] Website: http://www.pharmacology.unimelb.edu.au/statboss phone: +61 3 8344 7812 I'm interested an a modest statistics demonstration, but I can't figure out how do to the math myself. I'd like to have a few lines of code that produces a sequence of numbers. (whole numbers would probably be okay). I'd like to specify the number of numbers generated. Let's call that Z. I'd like also to specify the desired mean and standard deviation. I'd like the function (is this a function??) to work in such a way that if Z is large, the set of numbers generated, if graphed as a frequency distribution, would be normally distributed, i.e., Gaussian. If Z is rather small, then the mean and standard deviation of the numbers produced will would only approximate the desired mean and standard deviation. Different runs would produce different actual means and standard deviations. If Z is very small, like 3 or 4, the numbers will look almost random. ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Pricing / entry cost for this tool
For some time I have been toying with the idea that software should be sold on an income-weighted pricing scheme. If Richard can afford to pay more for Rev than Andre, it is in large part because he lives and earns in USA rather than Brazil. I have a couple of educational titles being sold by my University that cost the same number of Australian dollars to Harvard as they do to universities in Africa. It doesn't seem fair. Perhaps software prices could be adjusted for the average (modal) wage in a country. It wouldn't harm me for people in low wage countries to pay me almost nothing instead of absolutely nothing... On Nov 26, 2005, at 2:12 AM, Richard Gaskin wrote: Personally, I think Rev is priced too low. And Andre replied: Sh... don't talk that too loud, I am trying to sum some money to buy a new license and pounds are expensive ;-) Regards, Michael ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
Re: Pricing / entry cost for this tool
On 27/11/2005, at 9:13 AM, [EMAIL PROTECTED] wrote: I have a couple of educational titles being sold by my University that cost the same number of Australian dollars to Harvard as they do to universities in Africa. It doesn't seem fair. Perhaps software prices could be adjusted for the average (modal) wage in a country. It wouldn't harm me for people in low wage countries to pay me almost nothing instead of absolutely nothing... I don't currently make money from writing software - I make utilities for my own use. But if I did sell software I don't think I'd be interested in getting paid 3rd world wages (no offense intended - just don't know another way to say it) and paying U.S. rates for my housing, food etc. Making your software available at locally-affordable rates should not influence your ability to charge appropriate amounts in your country. Locally-affordable would be higher in the USA than in Brazil, and higher in Brazil than in Zaire. If you're rich and don't care, you can give your software away. I'm sure many of the pros on this list who make money programming, program because they like doing so. I'm sure the RunRev people love what they do. But we all have bills to pay too. When someone sets a price on a piece of software, I get to decide if that's worth my money. You've missed the point. If you only use a US-centric view of economic values then you leave out most of the people alive. What is offensive about asking a locally-affordable price for a product that has no cost for reproduction once produced? I don't figure it's their job to make sure I can afford it - they don't owe me a thing (which is the attitude that socialism breeds IMHO, sorry to digress into politics!). Given how we have gotten to the current world financial model, it would be more likely that we owe them than they owe us! Remember that the unequal distribution of wealth has come from the unequal distribution of power, not the unequal distribution of worth. It serves the ruling minority interests to equate means and success with moral worth, but we should be able to see beyond it. Certainly it's noble to want to see everyone have access to good software. But having directed a local soup kitchen for 6 years, I can tell you there are people in the U.S. who are desperately poor. Who would administrate a system that would charge based on income? I think I would always be VERY poor when I went to make my purchases! Marty Knapp I hope that you are kidding about that last comment, or at least exaggerating to illustrate a possible abuse of the proposed system. At the moment, in your country and mine, the very wealthy pay very little tax. The inevitability of abuse and loopholes should not be taken as a reason not to attempt to improve something. Regards, ___ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
