Revolution Analytics seems very confused
Revolution Analytics is a relaunch of a company previously known as REvolution Computing, built around the open source R language. Last week they sent around email claiming they were a new company (false), and asking for briefings in connection with an embargo this morning. I talked to Revolution Analytics yesterday, and they told me the embargo had been moved to Thursday.* However, Revolution apparently neglected to tell the press the same thing, and there’s an article out today — quoting me, because I’d given quotes in line with the original embargo, before I’d had the briefing myself. And what’s all this botched timing about? Mainly, it seems to be for a “statement of direction” about software Revolution Analytics hasn’t actually developed yet.
*More precisely, they spoke as if the embargo had been Thursday all along.
Anyhow, so far as I could tell, the Revolution Analytics story is something like:
- Revolution Analytics does statistical software. Any statistical software vendor that isn’t SAS (or SPSS) will be a lot cheaper than SAS, Revolution included.
- Revolution Analytics is adding closed-source extensions to the open source R project.
- In particular, Revolution Analytics has already added a visual development environment to R. I doubt it’s Eclipse-based, however, since Eclipse wasn’t mentioned once.
- Revolution Analytics also claims that R’s 4GL (Fourth-Generation Language) is a lot more useful than SAS’.
- A big knock on R is that you have to have your whole data set in RAM to use it. I didn’t gather that Revolution Analytics intends to fix this problem, but Revolution’s statement of direction includes something about parallel processing, so that you can use a cluster of machines at once. The Revolution folks were not in a position to usefully discuss how this will compare to, for example, Netezza’s implementation of R in TwinFin(i). (In fairness, Netezza hasn’t exactly released a lot of detail yet.)
- The R project, according to Revolution Analytics, boasts about 2000 contributors of statistical algorithmic extensions, from the investment community, from the investment and pharmaceutical industries, and so on.
- The R project, accordingly to Revolution Analytics, has about 20 core contributors, mainly academics. None works for Revolution, but one is on Revolution’s board.
- Last Fall, REvolution Computing fired half its employees, most of its management, and all of its board, in anticipation of a relaunch. Headquarters were also moved from Seattle to Silicon Valley, although development remains in Seattle.
- Revolution Analytics now has a little less than 30 people.
Comments
13 Responses to “Revolution Analytics seems very confused”
Leave a Reply
I don’t know very much about Revolution, but I do use R a lot. One of your points above isn’t true, namely it’s quite possible to use R on data sets larger than your RAM. You can address data that’s in a backing database (e.g. sqlite or a traditional DBMS), or use R to build mappers & reducers for Hadoop, and so on. Revolution themselves have done some good work in this area, some of which is available free.
[…] They probably need to hire more people now – Curt Monash, noted all-things-data software guru has the inside dope here […]
I’m one of the guys who created R. These guys are distributing GPLed software and not feeding anything back to the rights holders. Please don’t support them. Get the free version from the R-project instead and support the R Foundation.
Another flash in the pan, non-production stable, premature-to-market, money-hungry, get-rich-quick off a noble open source effort company with less thatn 30 people to suppor the globe. Sure.
Ross: can you be a bit more specific? There’s nothing wrong with “distributing GPLed software” so I’m guessing there’s something more that you’re not saying.
I can confirm it’s not Eclipse-based. I had thought it was but just downloaded the academic version and it is built on Visual Studio. From the about box:
Revolution R Enterprise powered by Visual Studio Shell.
Copyright © 2007 Microsoft Corporation.
Giving a free single-server version of all their closed-source code to academics != Revolution Analytics giving NOTHING back.
But it does give me pause when an open source commercialization company doesn’t have any core contributors as employees.
@Turadq,
Good find. Thanks!
Perhaps it’s not coincidental that Revolution development is in Seattle?
From the Revolution Analytics web site (http://revolutionanalytics.com/aboutus/leadership.php) it lists as a member of their board of directors:
“Robert Gentleman has been an innovator in the statistical computing world for more than 20 years. In 1996, he and fellow University of Aukland (NZ) professor Ross Ihaka released the “R” statistical computing language as a free software package after five years of development.”
It appears the two creators of R don’t agree on this topic.
@Ken,
But see Michael Wexler’s comment on http://www.dbms2.com/2010/05/07/in-database-sas-teradata-netezza-aster/
To what do you attribute this difference in perception?
Thanks,
CAM
@Curt,
Well, the way people *usually* use R is to import their data sets into in-memory data structures. They start loading bigger & bigger files this way, then hit memory limits, and say “R can’t handle big files”.
To handle big data requires different methods – see for example the ‘ff’, ‘biglm’, ‘bigglm’, ‘sqldf’, and in general anything on http://cran.r-project.org/web/views/HighPerformanceComputing.html .
@Ken,
Is there any serious downside to that stuff?
Hi there, just became aware of your blog through Google, and
found that it’s truly informative. I’m going to watch out for brussels.
I’ll appreciate if you continue this in future. Lots of people will be benefited from your writing.
Cheers!|
my web site: pit online 2013