Re: [R] [External Email] R Software Risk Analysis

2020-06-18 Thread Christopher W. Ryan
I use R every day with pretty sensitive data in my county health department. Of course, this is for manipulation and analysis of data pulled from their sources, not for interacting directly with, or updating, patient records in any clinically operational sense. As others have said, the structure an

Re: [R] R Software Risk Analysis

2020-06-18 Thread Richard O'Keefe
No, it was precisely my *point* that Matlab is proprietary. The medical researchers I knew a few years ago refused to use R on the grounds that the international agencies they dealt with all used SAS, which is proprietary. So I am wondering if "ALL programs that our employees/PROVIDERS use" includ

Re: [R] R Software Risk Analysis

2020-06-18 Thread Bert Gunter
Ummm...Except that Matlab is proprietary and for profit, not open source. Did you perhaps mean Octave? Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Thu, Jun

Re: [R] R Software Risk Analysis

2020-06-18 Thread Richard O'Keefe
Just as a matter of curiosity, what are some of the programs that have already been vetted, what methods were used, and how long did the vetting take? As the R guidance points out, R was not designed for creating or updating medical records, so it should be treated the same way as say LibreOffice

Re: [R] [External] Re: R Software Risk Analysis

2020-06-18 Thread Bert Gunter
As others have noted, R's vulnerabilities depend on the environments in which it is used. Perhaps the other issue is whether any downloaded R software could be problematic, perhaps due to malware. R's core functionality is, I'm sure fine. For the 20,000 or so packages on CRAN and elsewhere -- ?? On

Re: [R] [External] Re: R Software Risk Analysis

2020-06-18 Thread Richard M. Heiberger
You should start by reading R: Regulatory Compliance and Validation Issues: A guidance document for the use of R in regulated clinical trial environments. https://www.r-project.org/doc/R-FDA.pdf The official link to that file is at the R home page https://www.r-project.org/ In the left column, cli

Re: [R] R Software Risk Analysis

2020-06-18 Thread John Harrold
I work in Pharma and we use R in all the companies I've worked for. They are really paranoid and it's used in regulated environments as well with patient data. So there should be something they can do. Kristin: I can put you in touch with vendors who do our regulated work in R if you're interested

Re: [R] R Software Risk Analysis

2020-06-18 Thread David Winsemius
On 6/18/20 3:41 PM, John Harrold wrote: Hello Kristin, Are you talking about risk analysis from the perspective of software vulnerabilities? It appears that is exactly what is being asked. What is not clear is whether the installation would be offered to persons or groups on the network w

Re: [R] R Software Risk Analysis

2020-06-18 Thread Jeff Newmiller
R is open source software that is offered as-is, and many users of R utilize additional "contributed" packages which are developed and vetted independently of the R Core members. In addition, it is common for users of R to add minor functionality in the course of obtaining useful results, which

Re: [R] R Software Risk Analysis

2020-06-18 Thread John Harrold
Hello Kristin, Are you talking about risk analysis from the perspective of software vulnerabilities? John On Thu, Jun 18, 2020 at 3:21 PM Wait, Kristin wrote: > HI all, > > I am with a NYS major trauma center and all programs that our > employees/providers use must be vetted through the IT Dep

[R] R Software Risk Analysis

2020-06-18 Thread Wait, Kristin
HI all, I am with a NYS major trauma center and all programs that our employees/providers use must be vetted through the IT Department by way of a Risk Analysis. Is there someone I would talk to about this? I scoured your website and could not find a specific person. Thank you so much Kristin

Re: [R] comparing variances/distributions

2020-06-18 Thread Michael Dewey
Dear Ana This really depends on your scientific question. The two techniques you have shown do different things and there must be many more which could be applied. Michael On 17/06/2020 20:57, Ana Marija wrote: Hello, I have p values from two distributions, Pold and Pnew head(m) CHR

Re: [R] R 4.0.1/R-devel 2020-06-16-r78702 built with Intel Composer 19.1.1, error in R CMD make check on CentOS 7.7

2020-06-18 Thread Martin Maechler
> Ryan Novosielski > on Wed, 17 Jun 2020 22:04:30 + writes: > Same story with R-devel 2020-06-16-r78702, everything else the same. Should I be reporting this someplace else? Well, maybe Intel? {I've never heard of 'Intel Composer', and to me it does not look like Free / Ope