Re: [R] Rating R Helpers
Good morning, I read (and occasionally post to) R-Help through nabble.com (http://www.nabble.com/R-f13819.html) and there is a rating system there by post rather than by author. Message boards (e.g. www.boards.ie amongst many others) note on each message how many posts that a person has made since they joined the board. Best Regards, Sean Doran, Harold wrote: > > Since R is open source and help may come from varied levels of > experience on R-Help, I wonder if it might be helpful to construct a > method that can be used to "rate" those who provide help on this list. > > This is something that is done on other comp lists, like > http://www.experts-exchange.com/. > > I think some of the reasons for this are pretty transparent, but I > suppose one reason is that one could decide to implement the advise of > those with "superior" or "expert" levels. In other words, you can trust > the advice of someone who is more experienced more than someone who is > not. Currently, there is no way to discern who on this list is really an > R expert and who is not. Of course, there is R core, but most people > don't actually know who these people are (at least I surmise that to be > true). > > If this is potentially useful, maybe one way to begin the development of > such ratings is to allow the original poster to "rate" the level of help > from those who responded. Maybe something like a very simple > questionnaire on a likert-like scale that the original poster would > respond to upon receiving help which would lead to the accumulation of > points for the responders. Higher points would result in higher levels > of expertise (e.g., novice, ..., wizaRd). > > Just a random thought. What do others think? > > Harold > > > > > [[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > > -- View this message in context: http://www.nabble.com/Rating-R-Helpers-tf4925550.html#a14167605 Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
On Dec 5, 2007 12:49 AM, Patrick Connolly <[EMAIL PROTECTED]> wrote: > On Tue, 04-Dec-2007 at 05:32PM -0800, Ben Bolker wrote: > > |> > |> > |> > |> S Ellison wrote: > |> > > |> > Package review is a nice idea. But you raise a worrying point. > |> > Are any of the 'downright dangerous' packages on CRAN? > |> > If so, er... why? > |> > > |> > > |> <[EMAIL PROTECTED]> 12/01/07 7:21 AM >>> > |> >>I think the need for this is rather urgent, in fact. Most packages are > |> >>very good, but I regret to say some are pretty inefficient and others > |> >>downright dangerous. > |> > > |> > > |> > |> Presumably because the primary requirement for packages being > |> accepted on CRAN is that they pass "R CMD check". This is a fine > |> minimum standard -- it means that packages will definitely install -- > > That's not quite true. Package BRugs will go halfway through the > installation before a Linux user is given the information that it will > not work with Linux. The automated way the packages are listed > doesn't manage to collect that bit of information (and that's nothing > anyone should be ashamed of). > > Somewhere for adding information such as that could help avoid the > need for many people finding that out for themselves. > The bioconductor packages have the DESCRIPTION file's SystemRequirements field listed on the net so you can know what they are prior to downloading the file. For CRAN packages this information seems not to be shown on the net. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
On Tue, 04-Dec-2007 at 05:32PM -0800, Ben Bolker wrote: |> |> |> |> S Ellison wrote: |> > |> > Package review is a nice idea. But you raise a worrying point. |> > Are any of the 'downright dangerous' packages on CRAN? |> > If so, er... why? |> > |> > |> <[EMAIL PROTECTED]> 12/01/07 7:21 AM >>> |> >>I think the need for this is rather urgent, in fact. Most packages are |> >>very good, but I regret to say some are pretty inefficient and others |> >>downright dangerous. |> > |> > |> |> Presumably because the primary requirement for packages being |> accepted on CRAN is that they pass "R CMD check". This is a fine |> minimum standard -- it means that packages will definitely install -- That's not quite true. Package BRugs will go halfway through the installation before a Linux user is given the information that it will not work with Linux. The automated way the packages are listed doesn't manage to collect that bit of information (and that's nothing anyone should be ashamed of). Somewhere for adding information such as that could help avoid the need for many people finding that out for themselves. best -- ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~. ___Patrick Connolly {~._.~} Great minds discuss ideas _( Y )_Middle minds discuss events (:_~*~_:)Small minds discuss people (_)-(_) . Anon ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
John Sorkin wrote: > > I believe we need to know the following about packages: > (1) Does the package do what it purports to do, i.e. are the results > valid? > (2) Have the results generated by the package been validate against some > other statistical package, or hand-worked example? > (3) Are the methods used in the soundly based? > (4) Does the package documentation refer to referred papers or textbooks? > (5) In addition to the principle result, does the package return ancillary > values that allow for proper interpretation of the main result, (e.g. lm > gives estimates of the betas and their SEs, but also generates > residuals)?. > (6) Is the package easy to use, i.e. do the parameters used when invoking > the package chosen so as to allow the package to be flexible? > (7) Are the error messages produced by the package helpful? > (8) Does the package conform to standards of R coding and good programming > principles in general? > (9) Does the package interact will with the larger R environment, e.g. > does it have a plot method etc.? > (10) Is the package well documented internally, i.e. is the code easy to > follow, are the comments in the code adequate? > (11) Is the package well documented externally, i.e. through man pages and > perhaps other documentation (e.g. MASS and its associated textbook)? > > Numbers 1 to 3 are critical. The rest would be very nice to know (and should be part of a rating system), but in the end are more likely to lead to frustration than outright errors ... (i.e., you'll find out soon enough if a package is poorly documented, then you just won't use it). -- View this message in context: http://www.nabble.com/Rating-R-Helpers-tf4925550.html#a14163487 Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
S Ellison wrote: > > Package review is a nice idea. But you raise a worrying point. > Are any of the 'downright dangerous' packages on CRAN? > If so, er... why? > > <[EMAIL PROTECTED]> 12/01/07 7:21 AM >>> >>I think the need for this is rather urgent, in fact. Most packages are >>very good, but I regret to say some are pretty inefficient and others >>downright dangerous. > > Presumably because the primary requirement for packages being accepted on CRAN is that they pass "R CMD check". This is a fine minimum standard -- it means that packages will definitely install -- but there's nothing to stop anyone posting a package full of statistical nonsense to CRAN, as far as I know. I'm _not_ suggesting that R-core should take up this challenge, but this is where ratings come in. Ben Bolker -- View this message in context: http://www.nabble.com/Rating-R-Helpers-tf4925550.html#a14163486 Sent from the R help mailing list archive at Nabble.com. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
On Sun, 02-Dec-2007 at 11:20PM +, S Ellison wrote: |> Package review is a nice idea. But you raise a worrying point. |> Are any of the 'downright dangerous' packages on CRAN? Don't know about "dangerous", but I would like the opportunity to provide feedback or to have seen feedback from others when I contemplating using a package I see on CRAN. This is particularly true for package maintainers who seem to have vanished. Such a "rating" would have a natural place. I don't see where ratings for helpers would exist in cyberspace for enquirers to the list. A compulsory part of everyone's email signatures? Don't like that idea. -- ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~. ___Patrick Connolly {~._.~} Great minds discuss ideas _( Y )_Middle minds discuss events (:_~*~_:)Small minds discuss people (_)-(_) . Anon ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
"John Sorkin" <[EMAIL PROTECTED]> wrote: > I believe we need to know the following about packages: > (1) Does the package do what it purports to do, i.e. are the results valid? > (2) Have the results generated by the package been validate against some > other statistical package, or hand-worked example? > (3) Are the methods used in the soundly based? > (4) Does the package documentation refer to referred papers or textbooks? > (5) In addition to the principle result, does the package return ancillary > values that allow for proper interpretation of the main result, (e.g. lm > gives estimates of the betas and their SEs, but also generates residuals)?. > (6) Is the package easy to use, i.e. do the parameters used when invoking the > package chosen so as to allow the package to be flexible? > (7) Are the error messages produced by the package helpful? > (8) Does the package conform to standards of R coding and good programming > principles in general? > (9) Does the package interact will with the larger R environment, e.g. does > it have a plot method etc.? > (10) Is the package well documented internally, i.e. is the code easy to > follow, are the comments in the code adequate? > (11) Is the package well documented externally, i.e. through man pages and > perhaps other documentation (e.g. MASS and its associated textbook)? > > In addition to package evaluation and reviews, we also need some plan for the > future of R. Who will maintain, modify, and extend packages after the > principle author, or authors, retire? Software is never "done". Errors need > to be corrected, programs need to be modified to accommodate changes in > software and hardware. I have reasonable certainty that commercial software > (e.g. SAS) will be available in 10-years (and that PROC MIXED will still be a > part of SAS). I am far less sanguine about any number of R packages. > John Interesting questions. Re, the future : LaTeX provides an example. The more complex packages tend to stop developing when the original programmer loses interest. Sometimes another person picks one up, but not frequently. I think, for example, of the many slide-preparation packages, each more complex than the next, that have come and gone during my relatively short (15 yr) professional use of LaTeX. At its root, this is a rather deep question: how open-source, largely volunteer-developed software can survive over the long term, while continuing to improve and maintain high standards. We are rather early in the history of free software development to know the answer. -- Mike Prager, NOAA, Beaufort, NC * Opinions expressed are personal and not represented otherwise. * Any use of tradenames does not constitute a NOAA endorsement. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
I'm just a R user who joined the list searching solution for a problem. I do not think rating helpers is a good idea. For once, they do it freely; no need to harrash (?) anybody. On the other hand, it could have the opposite effect; people afraid to get a bad rating do not post their potentially valid answers. But more importantly, the list is full with examples of how to accomplish the same result with different approaches. Some might be more elegant than others, but for sure they show R potentialities. And I have seen several corrections/discussions among old timers themselves. Packages reviews are another issue. But if anybody is going through all that work, why not to make the appropriate corrections to the packages? They are GPL, aren't they? Best Pablo Mark Kimpel escribió: I'll throw one more idea into the mix. I agree with Bill that a rating system for respondents is probably not that practical and of not the highest importance. It also seems like a recipe for creating inter-personal problems that the list doesn't need. I do like Bill's idea of a review system for packages, which could be incorporated into my idea that follows... What I would find useful would be some sort of tagging system for messages. I can't count the times I've remembered seeing a message that addresses a question I have down the road but, when Googled, I can't find it. It would be so nice, for example, to reliably be able to find all messages related to a certain package or package function posted within the last X days. This could be implemented as simply as asking posters to provide keywords at the end of a message, but it would be great if they could somehow be pulled out of a message and stored in a DB. For instance keywords could be surrounded by a sequence of special characters, which a parser could then extract and store in a DB along with the message. Of course, this would be work to set up, but how many of our "experts" who so kindly give of their time, get exasperated when similar questions keep popping up on the list? Also, if we had a web-accessable DB, the responses, not the responders, could be rated as to how well a reply takes care of an issue. Thus, over time, a sort of auto-wiki could be born. I can think of more uses for this as well. For example a developer could quickly check to see what usability problems or suggestions have cropped up of on individual package. Mark On Dec 1, 2007 2:21 AM, [1]<[EMAIL PROTECTED]> wrote: This seems a little impractical to me. People respond so much at random and most only tackle questions with which they feel comfortable. As it's not a competition in any sense, it's going to be hard to rank people in any effective way. But suppose you succeed in doing so, then what? To me a much more urgent initiative is some kind of user online review system for packages, even something as simple as that used by Amazon.com has for customer review of books. I think the need for this is rather urgent, in fact. Most packages are very good, but I regret to say some are pretty inefficient and others downright dangerous. You don't want to discourage people from submitting their work to CRAN, but at the same time you do want some mechanism that allows users to relate their experience with it, good or bad. Bill Venables CSIRO Laboratories PO Box 120, Cleveland, 4163 AUSTRALIA Office Phone (email preferred): +61 7 3826 7251 Fax (if absolutely necessary): +61 7 3826 7304 Mobile: +61 4 8819 4402 Home Phone: +61 7 3286 7700 [2]mailto:[EMAIL PROTECTED] [3]http://www.cmis.csiro.au/bill.venables/ -Original Message- From: [EMAIL PROTECTED] [[5]mailto:[EMAIL PROTECTED] On Behalf Of Doran, Harold Sent: Saturday, 1 December 2007 6:13 AM To: R Help Subject: [R] Rating R Helpers Since R is open source and help may come from varied levels of experience on R-Help, I wonder if it might be helpful to construct a method that can be used to "rate" those who provide help on this list. This is something that is done on other comp lists, like [6]http://www.experts-exchange.com/. I think some of the reasons for this are pretty transparent, but I suppose one reason is that one could decide to implement the advise of those with "superior" or "expert" levels. In other words, you can trust the advice of someone who is more experienced more than someone who is not. Currently, there is no way to discern who on this list is really an R expert and who is not. Of course, there is R core, but most people don't actually know who these people are (at least I surmise that to be true). If this is potentially useful, maybe one way to begin the development of such ratings is to allow the original poster to "rate" the level of help from those who responded. Maybe something like a very simple questionnaire on a likert-like scale that the original poster would respond to upon receiving help which would lead to th
Re: [R] Rating R Helpers
I believe we need to know the following about packages: (1) Does the package do what it purports to do, i.e. are the results valid? (2) Have the results generated by the package been validate against some other statistical package, or hand-worked example? (3) Are the methods used in the soundly based? (4) Does the package documentation refer to referred papers or textbooks? (5) In addition to the principle result, does the package return ancillary values that allow for proper interpretation of the main result, (e.g. lm gives estimates of the betas and their SEs, but also generates residuals)?. (6) Is the package easy to use, i.e. do the parameters used when invoking the package chosen so as to allow the package to be flexible? (7) Are the error messages produced by the package helpful? (8) Does the package conform to standards of R coding and good programming principles in general? (9) Does the package interact will with the larger R environment, e.g. does it have a plot method etc.? (10) Is the package well documented internally, i.e. is the code easy to follow, are the comments in the code adequate? (11) Is the package well documented externally, i.e. through man pages and perhaps other documentation (e.g. MASS and its associated textbook)? In addition to package evaluation and reviews, we also need some plan for the future of R. Who will maintain, modify, and extend packages after the principle author, or authors, retire? Software is never "done". Errors need to be corrected, programs need to be modified to accommodate changes in software and hardware. I have reasonable certainty that commercial software (e.g. SAS) will be available in 10-years (and that PROC MIXED will still be a part of SAS). I am far less sanguine about any number of R packages. John John Sorkin M.D., Ph.D. Chief, Biostatistics and Informatics University of Maryland School of Medicine Division of Gerontology Baltimore VA Medical Center 10 North Greene Street GRECC (BT/18/GR) Baltimore, MD 21201-1524 (Phone) 410-605-7119 (Fax) 410-605-7913 (Please call phone number above prior to faxing) >>> <[EMAIL PROTECTED]> 12/1/2007 2:21 AM >>> This seems a little impractical to me. People respond so much at random and most only tackle questions with which they feel comfortable. As it's not a competition in any sense, it's going to be hard to rank people in any effective way. But suppose you succeed in doing so, then what? To me a much more urgent initiative is some kind of user online review system for packages, even something as simple as that used by Amazon.com has for customer review of books. I think the need for this is rather urgent, in fact. Most packages are very good, but I regret to say some are pretty inefficient and others downright dangerous. You don't want to discourage people from submitting their work to CRAN, but at the same time you do want some mechanism that allows users to relate their experience with it, good or bad. Bill Venables CSIRO Laboratories PO Box 120, Cleveland, 4163 AUSTRALIA Office Phone (email preferred): +61 7 3826 7251 Fax (if absolutely necessary): +61 7 3826 7304 Mobile: +61 4 8819 4402 Home Phone: +61 7 3286 7700 mailto:[EMAIL PROTECTED] http://www.cmis.csiro.au/bill.venables/ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Doran, Harold Sent: Saturday, 1 December 2007 6:13 AM To: R Help Subject: [R] Rating R Helpers Since R is open source and help may come from varied levels of experience on R-Help, I wonder if it might be helpful to construct a method that can be used to "rate" those who provide help on this list. This is something that is done on other comp lists, like http://www.experts-exchange.com/. I think some of the reasons for this are pretty transparent, but I suppose one reason is that one could decide to implement the advise of those with "superior" or "expert" levels. In other words, you can trust the advice of someone who is more experienced more than someone who is not. Currently, there is no way to discern who on this list is really an R expert and who is not. Of course, there is R core, but most people don't actually know who these people are (at least I surmise that to be true). If this is potentially useful, maybe one way to begin the development of such ratings is to allow the original poster to "rate" the level of help from those who responded. Maybe something like a very simple questionnaire on a likert-like scale that the original poster would respond to upon receiving help which would lead to the accumulation of points for the responders. Higher points would result in higher levels of expertise (e.g., novice, ..., wizaRd). Just a random thought. What do others think? Harold [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE
Re: [R] Rating R Helpers
Package review is a nice idea. But you raise a worrying point. Are any of the 'downright dangerous' packages on CRAN? If so, er... why? >>> <[EMAIL PROTECTED]> 12/01/07 7:21 AM >>> >I think the need for this is rather urgent, in fact. Most packages are >very good, but I regret to say some are pretty inefficient and others >downright dangerous. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
Mark Kimpel <[EMAIL PROTECTED]> [Sat, Dec 01, 2007 at 05:28:28PM CET]: > What I would find useful would be some sort of tagging system for messages. Hrm. I find tags immensely useful for entities which do not contain primarily text, such as photos. I am at doubt how keywords are important when they are not found in the text. There are situations where the first keyword that comes to mind is tiptoed around in the message, but I don't know if this is often the case. > I can't count the times I've remembered seeing a message that addresses a > question I have down the road but, when Googled, I can't find it. Is it a problem of way too many false positives or a problem of false negatives? Tags may help out in the second case, but in my experiencd it is rare. [...] > > Of course, this would be work to set up, but how many of our "experts" who > so kindly give of their time, get exasperated when similar questions keep > popping up on the list? Do you think that people who keep asking similar questions do so because they didn't do their homework first, or that they Googled and failed? -- Johannes H�sing There is something fascinating about science. One gets such wholesale returns of conjecture mailto:[EMAIL PROTECTED] from such a trifling investment of fact. http://derwisch.wikidot.com (Mark Twain, "Life on the Mississippi") __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
I don't think r-help is really intended for packages although for some very popular packages questions appear on it anyways sometimes. On Dec 1, 2007 11:28 AM, Mark Kimpel <[EMAIL PROTECTED]> wrote: > I'll throw one more idea into the mix. I agree with Bill that a rating > system for respondents is probably not that practical and of not the highest > importance. It also seems like a recipe for creating inter-personal problems > that the list doesn't need. > > I do like Bill's idea of a review system for packages, which could be > incorporated into my idea that follows... > > What I would find useful would be some sort of tagging system for messages. > I can't count the times I've remembered seeing a message that addresses a > question I have down the road but, when Googled, I can't find it. It would > be so nice, for example, to reliably be able to find all messages related to > a certain package or package function posted within the last X days. This > could be implemented as simply as asking posters to provide keywords at the > end of a message, but it would be great if they could somehow be pulled out > of a message and stored in a DB. For instance keywords could be surrounded > by a sequence of special characters, which a parser could then extract and > store in a DB along with the message. > > Of course, this would be work to set up, but how many of our "experts" who > so kindly give of their time, get exasperated when similar questions keep > popping up on the list? Also, if we had a web-accessable DB, the responses, > not the responders, could be rated as to how well a reply takes care of an > issue. Thus, over time, a sort of auto-wiki could be born. I can think of > more uses for this as well. For example a developer could quickly check to > see what usability problems or suggestions have cropped up of on individual > package. > > Mark > > On Dec 1, 2007 2:21 AM, <[EMAIL PROTECTED]> wrote: > > > > This seems a little impractical to me. People respond so much at random > > and most only tackle questions with which they feel comfortable. As > > it's not a competition in any sense, it's going to be hard to rank > > people in any effective way. But suppose you succeed in doing so, then > > what? > > > > To me a much more urgent initiative is some kind of user online review > > system for packages, even something as simple as that used by Amazon.com > > has for customer review of books. > > > > I think the need for this is rather urgent, in fact. Most packages are > > very good, but I regret to say some are pretty inefficient and others > > downright dangerous. You don't want to discourage people from > > submitting their work to CRAN, but at the same time you do want some > > mechanism that allows users to relate their experience with it, good or > > bad. > > > > > > Bill Venables > > CSIRO Laboratories > > PO Box 120, Cleveland, 4163 > > AUSTRALIA > > Office Phone (email preferred): +61 7 3826 7251 > > Fax (if absolutely necessary): +61 7 3826 7304 > > Mobile: +61 4 8819 4402 > > Home Phone: +61 7 3286 7700 > > mailto:[EMAIL PROTECTED] > > http://www.cmis.csiro.au/bill.venables/ > > > > -Original Message- > > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > > On Behalf Of Doran, Harold > > Sent: Saturday, 1 December 2007 6:13 AM > > To: R Help > > Subject: [R] Rating R Helpers > > > > Since R is open source and help may come from varied levels of > > experience on R-Help, I wonder if it might be helpful to construct a > > method that can be used to "rate" those who provide help on this list. > > > > This is something that is done on other comp lists, like > > http://www.experts-exchange.com/. > > > > I think some of the reasons for this are pretty transparent, but I > > suppose one reason is that one could decide to implement the advise of > > those with "superior" or "expert" levels. In other words, you can trust > > the advice of someone who is more experienced more than someone who is > > not. Currently, there is no way to discern who on this list is really an > > R expert and who is not. Of course, there is R core, but most people > > don't actually know who these people are (at least I surmise that to be > > true). > > > > If this is potentially useful, maybe one way to begin the development of > > such ratings is to allow the original poster to "rate" the level of help > > from those who responded. Maybe something like a very simple > > questionnaire on a likert-like scale that the original poster would > > respond to upon receiving help which would lead to the accumulation of > > points for the responders. Higher points would result in higher levels > > of expertise (e.g., novice, ..., wizaRd). > > > > Just a random thought. What do others think? > > > > Harold > > > > > > > > > >[[alternative HTML version deleted]] > > > > __ > > R-help@r-project.org mailing list > > https://stat.
Re: [R] Rating R Helpers
I'll throw one more idea into the mix. I agree with Bill that a rating system for respondents is probably not that practical and of not the highest importance. It also seems like a recipe for creating inter-personal problems that the list doesn't need. I do like Bill's idea of a review system for packages, which could be incorporated into my idea that follows... What I would find useful would be some sort of tagging system for messages. I can't count the times I've remembered seeing a message that addresses a question I have down the road but, when Googled, I can't find it. It would be so nice, for example, to reliably be able to find all messages related to a certain package or package function posted within the last X days. This could be implemented as simply as asking posters to provide keywords at the end of a message, but it would be great if they could somehow be pulled out of a message and stored in a DB. For instance keywords could be surrounded by a sequence of special characters, which a parser could then extract and store in a DB along with the message. Of course, this would be work to set up, but how many of our "experts" who so kindly give of their time, get exasperated when similar questions keep popping up on the list? Also, if we had a web-accessable DB, the responses, not the responders, could be rated as to how well a reply takes care of an issue. Thus, over time, a sort of auto-wiki could be born. I can think of more uses for this as well. For example a developer could quickly check to see what usability problems or suggestions have cropped up of on individual package. Mark On Dec 1, 2007 2:21 AM, <[EMAIL PROTECTED]> wrote: > This seems a little impractical to me. People respond so much at random > and most only tackle questions with which they feel comfortable. As > it's not a competition in any sense, it's going to be hard to rank > people in any effective way. But suppose you succeed in doing so, then > what? > > To me a much more urgent initiative is some kind of user online review > system for packages, even something as simple as that used by Amazon.com > has for customer review of books. > > I think the need for this is rather urgent, in fact. Most packages are > very good, but I regret to say some are pretty inefficient and others > downright dangerous. You don't want to discourage people from > submitting their work to CRAN, but at the same time you do want some > mechanism that allows users to relate their experience with it, good or > bad. > > > Bill Venables > CSIRO Laboratories > PO Box 120, Cleveland, 4163 > AUSTRALIA > Office Phone (email preferred): +61 7 3826 7251 > Fax (if absolutely necessary): +61 7 3826 7304 > Mobile: +61 4 8819 4402 > Home Phone: +61 7 3286 7700 > mailto:[EMAIL PROTECTED] > http://www.cmis.csiro.au/bill.venables/ > > -Original Message- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > On Behalf Of Doran, Harold > Sent: Saturday, 1 December 2007 6:13 AM > To: R Help > Subject: [R] Rating R Helpers > > Since R is open source and help may come from varied levels of > experience on R-Help, I wonder if it might be helpful to construct a > method that can be used to "rate" those who provide help on this list. > > This is something that is done on other comp lists, like > http://www.experts-exchange.com/. > > I think some of the reasons for this are pretty transparent, but I > suppose one reason is that one could decide to implement the advise of > those with "superior" or "expert" levels. In other words, you can trust > the advice of someone who is more experienced more than someone who is > not. Currently, there is no way to discern who on this list is really an > R expert and who is not. Of course, there is R core, but most people > don't actually know who these people are (at least I surmise that to be > true). > > If this is potentially useful, maybe one way to begin the development of > such ratings is to allow the original poster to "rate" the level of help > from those who responded. Maybe something like a very simple > questionnaire on a likert-like scale that the original poster would > respond to upon receiving help which would lead to the accumulation of > points for the responders. Higher points would result in higher levels > of expertise (e.g., novice, ..., wizaRd). > > Just a random thought. What do others think? > > Harold > > > > >[[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and p
Re: [R] Rating R Helpers
On Dec 1, 2007 2:21 AM, <[EMAIL PROTECTED]> wrote: > To me a much more urgent initiative is some kind of user online review > system for packages, even something as simple as that used by Amazon.com > has for customer review of books. > > I think the need for this is rather urgent, in fact. Most packages are > very good, but I regret to say some are pretty inefficient and others > downright dangerous. You don't want to discourage people from > submitting their work to CRAN, but at the same time you do want some > mechanism that allows users to relate their experience with it, good or > bad. You can get a very rough idea of this automatically by making a list of which packages are dependents of other packages. library(pkgDepTools) # from BioC AP <- available.packages() dep <- AP[, "Depends"] deps <- unlist(sapply(dep, pkgDepTools:::cleanPkgField)) sort(table(deps)) Of course some packages are more naturally end-user oriented and so would never make such a list and others may be good but just no one knows about them so they have never been used. Some packages might get on the list because an author has two packages and one uses the other so an enhancement could be to eliminate dependencies with a common author. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
This seems a little impractical to me. People respond so much at random and most only tackle questions with which they feel comfortable. As it's not a competition in any sense, it's going to be hard to rank people in any effective way. But suppose you succeed in doing so, then what? To me a much more urgent initiative is some kind of user online review system for packages, even something as simple as that used by Amazon.com has for customer review of books. I think the need for this is rather urgent, in fact. Most packages are very good, but I regret to say some are pretty inefficient and others downright dangerous. You don't want to discourage people from submitting their work to CRAN, but at the same time you do want some mechanism that allows users to relate their experience with it, good or bad. Bill Venables CSIRO Laboratories PO Box 120, Cleveland, 4163 AUSTRALIA Office Phone (email preferred): +61 7 3826 7251 Fax (if absolutely necessary): +61 7 3826 7304 Mobile: +61 4 8819 4402 Home Phone: +61 7 3286 7700 mailto:[EMAIL PROTECTED] http://www.cmis.csiro.au/bill.venables/ -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Doran, Harold Sent: Saturday, 1 December 2007 6:13 AM To: R Help Subject: [R] Rating R Helpers Since R is open source and help may come from varied levels of experience on R-Help, I wonder if it might be helpful to construct a method that can be used to "rate" those who provide help on this list. This is something that is done on other comp lists, like http://www.experts-exchange.com/. I think some of the reasons for this are pretty transparent, but I suppose one reason is that one could decide to implement the advise of those with "superior" or "expert" levels. In other words, you can trust the advice of someone who is more experienced more than someone who is not. Currently, there is no way to discern who on this list is really an R expert and who is not. Of course, there is R core, but most people don't actually know who these people are (at least I surmise that to be true). If this is potentially useful, maybe one way to begin the development of such ratings is to allow the original poster to "rate" the level of help from those who responded. Maybe something like a very simple questionnaire on a likert-like scale that the original poster would respond to upon receiving help which would lead to the accumulation of points for the responders. Higher points would result in higher levels of expertise (e.g., novice, ..., wizaRd). Just a random thought. What do others think? Harold [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
Also just MHO, but I think it is a good idea, if for no other reason than it creates an additional incentive for people to respond and to do so in a thoughtful way. I think the rating systems on some other boards, and those used at commercial sites such as Amazon, testify to the fact that people often do care about their ratings, and this, in turn, tends to bring the discourse to a higher level. Of course, it would also help newbies get some idea about how to weight responses - something that takes quite a long time otherwise (longer than 3 months I'd argue). However, I'm not sure that the ratings should be restricted to only those who asked the question given that it isn't only the questioner who has a valid opinion on responses. Why not open it to everyone (and increase sample sizes)? Mark - I agree that the quality of the ratings themselves is going to depend on the questioner/raters. But the noise will wash out over time, or in stats speak, your mean rating should be unbiased estimate of your response helpfulness and its variance should decrease as more people rate your responses :). I guess that assumes independence though... In any event, I'd love to see it happen, Matt On Nov 30, 2007 1:53 PM, <[EMAIL PROTECTED]> wrote: > >From: "Doran, Harold" <[EMAIL PROTECTED]> > >Date: 2007/11/30 Fri PM 02:12:36 CST > >To: R Help <[EMAIL PROTECTED]> > >Subject: [R] Rating R Helpers > > Of course, it's just MHO, but I think your suggestion is more work than > neccesary because it becomes pretty obvious who the R-experts are once you've > been on the list for more than 3 months. > > Also, the rating system wouldn't work well > because the quality of the rating would depend > on the experience of the person who asked the > question. A beginner might think a particular answer > is good while a more experienced person might think > it was bad. > >Mark > > > > > > > > > > > > > > > > >Since R is open source and help may come from varied levels of > >experience on R-Help, I wonder if it might be helpful to construct a > >method that can be used to "rate" those who provide help on this list. > > > >This is something that is done on other comp lists, like > >http://www.experts-exchange.com/. > > > >I think some of the reasons for this are pretty transparent, but I > >suppose one reason is that one could decide to implement the advise of > >those with "superior" or "expert" levels. In other words, you can trust > >the advice of someone who is more experienced more than someone who is > >not. Currently, there is no way to discern who on this list is really an > >R expert and who is not. Of course, there is R core, but most people > >don't actually know who these people are (at least I surmise that to be > >true). > > > >If this is potentially useful, maybe one way to begin the development of > >such ratings is to allow the original poster to "rate" the level of help > >from those who responded. Maybe something like a very simple > >questionnaire on a likert-like scale that the original poster would > >respond to upon receiving help which would lead to the accumulation of > >points for the responders. Higher points would result in higher levels > >of expertise (e.g., novice, ..., wizaRd). > > > >Just a random thought. What do others think? > > > >Harold > > > > > > > > > > [[alternative HTML version deleted]] > > > >__ > >R-help@r-project.org mailing list > >https://stat.ethz.ch/mailman/listinfo/r-help > >PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > >and provide commented, minimal, self-contained, reproducible code. > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- Matthew C Keller Asst. Professor of Psychology University of Colorado at Boulder www.matthewckeller.com __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
Maybe I haven't been on the list long enough, but how bad is the R advice on this list? Is a rating system even necessary? As Mark Leeds pointed out, there is the issue that what a beginner thinks is good advice isn't what an expert thinks in good advice. There is also the converse, expert advice is sometimes quite confusing to beginners, but perfectly sensible to other experts. Doran, Harold brought next idea : > Since R is open source and help may come from varied levels of > experience on R-Help, I wonder if it might be helpful to construct a > method that can be used to "rate" those who provide help on this list. > > This is something that is done on other comp lists, like > http://www.experts-exchange.com/. > > I think some of the reasons for this are pretty transparent, but I > suppose one reason is that one could decide to implement the advise of > those with "superior" or "expert" levels. In other words, you can trust > the advice of someone who is more experienced more than someone who is > not. Currently, there is no way to discern who on this list is really an > R expert and who is not. Of course, there is R core, but most people > don't actually know who these people are (at least I surmise that to be > true). > > If this is potentially useful, maybe one way to begin the development of > such ratings is to allow the original poster to "rate" the level of help > from those who responded. Maybe something like a very simple > questionnaire on a likert-like scale that the original poster would > respond to upon receiving help which would lead to the accumulation of > points for the responders. Higher points would result in higher levels > of expertise (e.g., novice, ..., wizaRd). > > Just a random thought. What do others think? > > Harold > > > > > [[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
>From: "Doran, Harold" <[EMAIL PROTECTED]> >Date: 2007/11/30 Fri PM 02:12:36 CST >To: R Help <[EMAIL PROTECTED]> >Subject: [R] Rating R Helpers Of course, it's just MHO, but I think your suggestion is more work than neccesary because it becomes pretty obvious who the R-experts are once you've been on the list for more than 3 months. Also, the rating system wouldn't work well because the quality of the rating would depend on the experience of the person who asked the question. A beginner might think a particular answer is good while a more experienced person might think it was bad. Mark >Since R is open source and help may come from varied levels of >experience on R-Help, I wonder if it might be helpful to construct a >method that can be used to "rate" those who provide help on this list. > >This is something that is done on other comp lists, like >http://www.experts-exchange.com/. > >I think some of the reasons for this are pretty transparent, but I >suppose one reason is that one could decide to implement the advise of >those with "superior" or "expert" levels. In other words, you can trust >the advice of someone who is more experienced more than someone who is >not. Currently, there is no way to discern who on this list is really an >R expert and who is not. Of course, there is R core, but most people >don't actually know who these people are (at least I surmise that to be >true). > >If this is potentially useful, maybe one way to begin the development of >such ratings is to allow the original poster to "rate" the level of help >from those who responded. Maybe something like a very simple >questionnaire on a likert-like scale that the original poster would >respond to upon receiving help which would lead to the accumulation of >points for the responders. Higher points would result in higher levels >of expertise (e.g., novice, ..., wizaRd). > >Just a random thought. What do others think? > >Harold > > > > > [[alternative HTML version deleted]] > >__ >R-help@r-project.org mailing list >https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide http://www.R-project.org/posting-guide.html >and provide commented, minimal, self-contained, reproducible code. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
A nice model might be http://www.workingwithrails.com/. It also suggests other possible measures of authority: * member of R core * packages written by/contributed to * R conference attended/presented at * contributer to wiki Hadley On 11/30/07, Doran, Harold <[EMAIL PROTECTED]> wrote: > Since R is open source and help may come from varied levels of > experience on R-Help, I wonder if it might be helpful to construct a > method that can be used to "rate" those who provide help on this list. > > This is something that is done on other comp lists, like > http://www.experts-exchange.com/. > > I think some of the reasons for this are pretty transparent, but I > suppose one reason is that one could decide to implement the advise of > those with "superior" or "expert" levels. In other words, you can trust > the advice of someone who is more experienced more than someone who is > not. Currently, there is no way to discern who on this list is really an > R expert and who is not. Of course, there is R core, but most people > don't actually know who these people are (at least I surmise that to be > true). > > If this is potentially useful, maybe one way to begin the development of > such ratings is to allow the original poster to "rate" the level of help > from those who responded. Maybe something like a very simple > questionnaire on a likert-like scale that the original poster would > respond to upon receiving help which would lead to the accumulation of > points for the responders. Higher points would result in higher levels > of expertise (e.g., novice, ..., wizaRd). > > Just a random thought. What do others think? > > Harold > > > > > [[alternative HTML version deleted]] > > __ > R-help@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. > -- http://had.co.nz/ __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] Rating R Helpers
Doran, Harold wrote: > Since R is open source and help may come from varied levels of > experience on R-Help, I wonder if it might be helpful to construct a > method that can be used to "rate" those who provide help on this list. > > This is something that is done on other comp lists, like > http://www.experts-exchange.com/. > > I think some of the reasons for this are pretty transparent, but I > suppose one reason is that one could decide to implement the advise of > those with "superior" or "expert" levels. In other words, you can trust > the advice of someone who is more experienced more than someone who is > not. Currently, there is no way to discern who on this list is really an > R expert and who is not. Of course, there is R core, but most people > don't actually know who these people are (at least I surmise that to be > true). > > If this is potentially useful, maybe one way to begin the development of > such ratings is to allow the original poster to "rate" the level of help > from those who responded. Maybe something like a very simple > questionnaire on a likert-like scale that the original poster would > respond to upon receiving help which would lead to the accumulation of > points for the responders. Higher points would result in higher levels > of expertise (e.g., novice, ..., wizaRd). > > Just a random thought. What do others think? > > Harold > Not necessary a bad idea but how do you take into account how much each responder is paid to respond? Frank -- Frank E Harrell Jr Professor and Chair School of Medicine Department of Biostatistics Vanderbilt University __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.