Apr 30, 2015

Failure to Launch

Five months ago, I requested an invite to IBM Verse. I was hoping to demo the features they were heralding, and see how they all came together into what IBM was touting as the future of email.

I finally got my invite.

Here's an account of my experience getting up and running on IBM's new platform.

6:02 PM - IBM sends me an email inviting me to create my account. Yes! Let's take a look at this nirvana of email.
"Delightful". That word... I do not think it means what you think it means.
6:16 PM - I click the link and provide some basic information, and receive a message that my account is being created. I will get another email when it is ready. Within the next business day. Huh? Seems a little slow, but okay.

6:26 PM - IBM sends me another email. My account is almost ready! Nice! In only ten minutes, after setting my expectation for 24 hours. I just need to set a password. Which I probably could have done in the first step. But let's set that password!

7:36 PM - I click the link to set my password. Success! Sort of. They are in the process of setting my password. Not sure why it's not instant. Perhaps it's going on the block chain? Although I really hope not, it's one of my "good" passwords and we don't need to make a permanent record of that.

7:42 PM - Another email... my account is ready! Really ready this time. "Get ready for mail that understands you, for less clutter and more clarity, for connecting you to the people who matter most." Yes! I am ready!

7:59 PM - I sign in! And wait while the app loads... there are FOUR different loading interstitials, done in what I would call an "early Netscape-era motif":
  • Please wait...
  • Please wait, signing on...
  • Loading...
  • Loading IBM Verse...
Didn't expect that, but it must be loading something amazing! And then IBM Verse is loaded. I don't know if the site was engineered to play a desolate wind sound, or if I just imagine it. But I'm in, and I'm sure I'll have emails there in no time!
"Happiness is an empty inbox." Indeed.
8:05 PM - I set up my profile picture, take a look around, and quickly realize there's no way to link other email accounts. I will need to use my new IBM Verse email address in order to get any emails.

8:40 PM - After distracting myself with Coursera for a while, I head to the IBM Verse forums to confirm that there really is no way to link other email accounts. And there is no way. Unless I'm on a paid version of Verse.

As other people have said:
  • Agreed ... and confused: "Possibly I mistakenly thought VERSE was all about email consolidation."
  • Worst marketing strategy: "This is the worst marketing strategy ever. You tease us with a product that you claim is going to change the way email is used and then we cannot evaluate it fully unless we buy the full version. IBM, you have lost a potential customer. I will not even attempt to get my organization to consider verse."
9:44 PM - I realize I have never been more underwhelmed by a product launch. I go back to the site and take one last look at my inbox. The wind howls a bit louder. I realize that if "Happiness is an empty inbox" then I am going to be absolutely ecstatic with IBM Verse.

Feb 23, 2015

RightClickR!

After I took the Statistics One and Data Analysis series courses on Coursera last year, I told myself that I was going to stop doing statistical analysis in Excel. I had a good intro to R, and found that it made a lot of the analysis I did in Excel--from basic statistical analysis and exploratory visualization--extremely easy and quick. And it was reproducible.

But then I got lazy. I stopped exporting Excel files to CSV. I stopped loading those CSVs into R. And because of those two extra steps that I stopped doing, everything I learned in R became a little hazy.

There was really only one barrier to doing more in R and keeping my R skills fresh--getting those spreadsheets into R. So I quickly built RightClickR, a registry setting / batch file combo that lets me right-click on an Excel file and have it open up in R. All tabs get loaded into a dataframe in the format of xlsx.<tab name>. As I write that, I realized I should probably do some tab name cleansing, but that's a project for another day. Maybe Thursday.

Now just to get me some data!

As you might notice in the screenshot above, using RightClickR creates an R file corresponding to the Excel spreadsheet. This file will continue to be loaded each time you use RightClickR for that spreadsheet, and as a result is a good place to drop any munging you need to do on your data set.

If you're handy with batch / registry files and you spend time in Excel that you'd rather spend in R, maybe the files below will help you.

RightClickR.bat

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
@echo off
set RCR_RFILE=%~n1.R
set RCR_FILEPATH=%1
set RCR_PATH=%~p1
set RCR_NEWPATH=%RCR_FILEPATH:\=/%
if exist %RCR_PATH%%RCR_RFILE% goto launchR
echo library(xlsx) > %RCR_RFILE%
echo rcr_file ^<- '%RCR_NEWPATH%' >> %RCR_RFILE%
echo rcr_wb ^<- loadWorkbook(rcr_file) >> %RCR_RFILE%
echo rcr_sheets ^<- getSheets(rcr_wb) >> %RCR_RFILE%
echo for (rcr_s in 1:length(rcr_sheets)) { >> %RCR_RFILE%
echo   rcr_data ^<- read.xlsx(rcr_file,rcr_s) >> %RCR_RFILE%
echo   assign(paste("xlsx.",names(rcr_sheets[rcr_s]),sep=""),rcr_data) >> %RCR_RFILE%
echo } >> %RCR_RFILE%
echo remove(rcr_data) >> %RCR_RFILE%
:launchR
start C:\Progra~1\RStudio\bin\rstudio.exe %RCR_PATH%%RCR_RFILE%
:exit

And here's the reg file. The usual disclaimers about modifying your registry apply--if you don't know what you're doing you can bork things up, so only do this if you have done this before and are comfortable with making registry changes.

RightClickR.reg

1
2
3
4
5
6
Windows Registry Editor Version 5.00

[HKEY_CLASSES_ROOT\Excel.Sheet.12\shell\OpenInR]

[HKEY_CLASSES_ROOT\Excel.Sheet.12\shell\OpenInR\command]
@="C:\\\\Users\\\\bhayek\\\\Desktop\\\\Misc\\\\Personal\\\\Data\\\\RightClickR\\\\RightClickR.bat %1"

To adapt this for your setup:

  1. Install the xlsx package in R if you haven't already
  2. Change line 17 in RightClickR.bat to reflect the path to RStudio on your local machine
  3. Update line 6 in RightClickR.reg to reflect the path where you have saved the batch file; note that you will need to double-double-escape your backslashes; each backslash in your path will need to be replaced by four for the registry file
  4. Double-click RightClickR.reg file to make the registry changes and set up your right-click shortcut to R!