« A first look at Spark | Main | Because it's Friday: A straight edge, and a compass »

January 23, 2015

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Hi David,

That's exciting news! Can you comment on the integration of Revo R into Excel or SQL Server?

Cheers,
Bob

Hi Bob, we don't have any specific details on integration plans yet, but we'll share here when we do. And thanks!

Yeeks! Just don't fuck it up.

What a joke. You're really working hard to try and convince readers that this is a good match, going on and on about how supportive Microsoft is of open-source. You were probably sweating while trying to come up with excuses as to why this is good, knowing that you were typing bullshit. I would suggest growing a pair of balls and just being honest, but I'm sure you've never had to do that in your career.

This is super exciting!
As a CTO from a recently acquired HPC Cloud company (GreenButton) into Azure with a similarly broad range of customers and open-source technology, I can assure you you've made a good decision and Revolution Analytics and your customers are in great hands. There's no better time to be part of Microsoft and it's an incredibly exciting time to be working for the company!

So welcome. I'm sure there'll be opportunities for us to work together (I'm on the Azure Big Compute team). See you soon!

Anonymous, I've never been anything but frank on this blog and this is no exception. I'm truly excited for the future, and I'm sure I speak for the rest of the team as well.

Awesome news!! Just started developing some analytics to be deployed on Azure. Was thinking of using RRE on HDInsights. Super excited to hear this news.

When Microsoft wiil offer a R version to Surface ?

Anyone else find it funny that the guy telling David to grow a pair of balls refused to sign his name to his post?

Excellent news, extend R to a larger user community

WOW!
Congratulations David! I wish you and others in REvoluion a lot of luck and success with this major change.

With regards,
Tal

Thanks Tal! And thanks for all the great work on r-bloggers.com. Such a service to the community.

Well, grats, I guess... But why? Is there any good reason besides the money? Just remember what Microsoft did to Nokia...

I just hope that at least you will pay more attention to OpenSUSE now since Novell de facto belongs to Microsoft too.

Not happy - Part of R's charm was that it was unemcumbered with a large Corporate problems. Time to market and market share will drive it from now on.

Congrats David. This is a great outcome. I think the haters are always gonna hate, but the MSFT of 2015 is quite different from that of the past. They are embracing cloud, open standards and even open source. Lets see what this team can continue to deliver!
--Zack

everybody has their price. good luck revolution, see you on the way down. microsoft is dead in the water after the total failure of surface, windows mobile, the social pariah that is windows 8.1, and windows 10 that's no more than a service pack which they can't even give away. nothing kills an organization of people like giant corporate machinery; watch revolution's more free thinking engineers jump ship when the mba's move in with their value statements and strategic thrusts.

Traditional R uses were even suspicious of you before you become part of the Borg.


If the Bong was so committed to open source will they now make an MS Office version for Linux?

Hmm, I have mixed feelings, too. If what you @David write about MS really were true, I'd be happy; then,
I feel similar to Tim Washburn's(Jan.23, 18:02).

David can you please make clear that this is about Revo, and not about R.
See what others make from the MS blog: "Micrsoft buys Revo ... and R" : http://bit.ly/15EVa2N

Not funny! I hope you can ensure that such misleading info is not spread, but corrected.

Martin Maechler, ETH Zurich,
and R Core team since its beginning in 1995

I have very mixed feelings as well. From companies perspective - sure they probably got nice pay day, which is nice.

I doubt anybody saw Revo as very open source company, but still they contributed very well to community, R packages and R popularity. Hope some of these will keep living since it will be beneficial to Microsoft as well.

The only real point of M$ buying RA is to integrate SS with RA in a more robust, and user friendly way, than Joe Conway's PL/R does for Postgres. Joe's done that all on his own, so far as I know, and works well. All those big brains in Redmond should be able to make a truly seamless integration. Run stats and graphs along with regular SQL all from the database. I've done it with PG and PL/R. It's cool.

Running SS from an R session (RA version or real R) has been standard for years, so there's not much more to offer in such connectivity. Unless M$ is just going to blow smoke. Not that they'd ever do that, of course.

@Martin is absolutely right, this is about MSFT acquiring Revolution Analytics, not R. R will continue as it's always done as an OSS project (which we'll continue to support). So statements like "Time to market and market share will drive it from now on" don't make sense. Martin, I'll reach out to the author of that article and see if we will correct that misleading headline.

David -- nice article and coverage. I have been a professional consultant using SAS for years, and am now a Microsoft MVP, largely based on my knowledge of SSAS and promotion of the SSAS/Excel functions. The open source mentions you have are important since we are no longer under Ballmer's vision: today .NET is open source.

I continue to have people interested in R, which I view as a mature technology at v3. Several hundred people saw my R presentation in Chicago 2013: I mentioned Revolution Analytics in the deck. I also talked about where I believe Julia fits:
http://www.slideshare.net/marktab/data-analysis-with-r-and-julia-201305

As a research analyst for Gigaom Research, I published a paper last fall looking at machine learning technologies (both commercial and open source). I gave some coverage to R and Revolution Analytics in that paper:
http://research.gigaom.com/analyst/mark-tabladillo/

There's a lot of interesting machine learning activity at Microsoft, including Microsoft Research: though lately, Joseph Sirosh seasoned at Amazon is at the center of many new announcements. I just signed up to lead a user group through the PASS organization (free, user run, and people may want to join if they want to tap into the many free resources the community provides). This virtual online chapter is called PASS Data Science: We are in the process of establishing the portal, but already have several people wanting to speak. We hope to have events with attendance at 200+ (based on similar group attendance numbers). The mission of the group is Microsoft Data Science technologies, and other open-source or commercial solutions integrated with Microsoft technologies (such as Azure).

As Revolution Analytics technology leaders bring their vision into Microsoft, professionals who are interested in the free spread of helpful information (as I have been for both the SAS and Microsoft communities) will find many opportunities in the Microsoft ecosystem. Microsoft offers discounts to educational and non-profit institutions; students can find access at dreamspark.com; new startups should leverage BizSpark.com

I encourage people to connect with me on LinkedIn or Twitter @marktabnet

I read the newsletter and was disappointed.

This is bad news for Linux stack customers.

Customer first? No way. Are prices going to decrease? Doubtful. Is the technical support model going to stay efficient and friendly? Doubtful. Pity to the poor contract managers in companies who are Linux stack who now have to negotiate with Microsoft. The timing is negative for Big Data for the Enterprise in R.

Possibly the worst outcome is that Microsoft will spike the Linux distributions of Revolutions Analytics software. Not further develop in Linux and maybe sell the code eventually, focusing development solely on Windows distributions in the meantime. It’d be a competitive advantage for them to do so. But, there is time. Revolution Analytics software is well developed as far as what it can do and the value it can add. It works and is solid. Customers can use the releases already in the pipeline to take a wait and see approach and to start thinking about a transition to other software.

A big winner in this acquisition is the Microsoft stack customer base as they are sure to receive integration of the code into their stack. Microsoft missed the first few years of the shift to analytics and this acquisition is one way to meet expectations and needs of their customers. The Python community will benefit as Big Data for the Enterprise in R starts to consider alternatives. (I don’t think the open source R community will notice or be impacted.)

This is Microsoft’s to lose. They have a culture of software that doesn’t play well with others and that gets dysfunctional over time. They have a history of hating Linux, which, admittedly, they are in the process of trying to change. I just don’t see how they can do a good job with Linux-based, analytics software. The best thing Microsoft can do is treat the Linux distributions of Revolution Analytics software as a first-class citizen and then maybe this can work. They have to be fair to Linux stack customers. To continue to attract companies to Big Data in the Enterprise with R, they have to be transparent and open about their intentions.

@Philip -- thanks for your insights.

Periodically, Microsoft will ask me (among many people) what I think about their analytics technologies. David put the photo of the Linux support, but I wonder what their general plans are too. For sure: Microsoft is motivated to completely utilize and expand the already-existing physical facilities which support Azure.

Here is a video 5 minute tour of an Azure data center: https://www.youtube.com/watch?v=zXsoygN_v7A

On your claim about Microsoft missing the first few years: Microsoft introduced several robust analytics algorithms with SQL Server 2000 (the first vendor to do so at the database), with a more full delivery in 2005.

I believe Microsoft's marketing focus in 2000-2005 was the core virtues of SQL Server altogether rather than the analytics technologies. The entire SQL Sever stack includes OLTP, DW, Tabular, OLAP. Put into more direct terms: they did not market the analytics portion as well as it could have been. Key people who helped develop that data mining technology left Microsoft and formed Predixion Software in 2009: they have a expanding client base with their solid Microsoft-based cloud service.

Microsoft's analytics future will be jointly decided by the experts at Revolution Analytics and the current Microsoft team. Joseph Sirosh is relatively new in his position, and his whole team is new too (some have a lot of years with Microsoft but in other areas). Several key Revolution Analytics people have a new experience base to add, including from Accenture and from SAS.

You mentioned Python: a few months ago, Joseph Sirosh's team adopted "Python Tools for Visual Studio". Python is one of three languages (along with R and C#) which have been already integrated into Azure ML. In the future, I'm hoping to see Python scripting incorporated into Azure ML: already R scripting is available. http://pytools.codeplex.com/

I agree with your observation that the typical desktop R user will not see any differences. This merger is about what could be done for the large-scale commercial users. Many data scientists trained in the last 10 years know R and they know it very well (what happened to SAS Enterprise Miner especially for students is off topic). The Azure ML free tier might appeal to the smaller scale R projects too, especially students. http://azure.microsoft.com/en-us/updates/azure-machine-learning-now-offers-free-usage-tier/

Finally, I offer my recommendations for how to directly "get in the know": here's how I know. Joseph Sirosh has put a priority on his team's Machine Learning blog (each division of Microsoft has this ability, but some blogs are more active than others): http://blogs.technet.com/b/machinelearning/ The general Twitter account is @MLatMSFT

More formally, I recommend commercial organizations and consultants to become part of Microsoft's Partner Network. Companies such as Accenture are members. The Partner Network is the path to meetings only for partners, such as the annual Worldwide Partner Conference. Last year, I went to a few free day-long sessions on Azure. Also, partners have the opportunity to sign an NDA with Microsoft allowing them to participate in new product development (by topic). https://mspartner.microsoft.com/en/us/Pages/index.aspx

I have an NDA relationship because I am part of the Microsoft MVP program: I am among the hundreds of potential people that Microsoft might ask about their product technologies. Some of these discussions happen on Yammer. Whether Microsoft asks for feedback or participation is entirely at their discretion, and largely reflects how different product teams approach their work.

Microsoft 2015 is generally motivated to work with people who have organizations ready to take their latest-and-greatest technologies to production, even and especially when that technology is in preview. They especially want people who can take their technology to the performance limits, and provide quantitative technical feedback.

Independent of the NDA-world, Microsoft monitors feedback and bug reports for selected technologies at http://connect.microsoft.com/ If you choose to post something there, you have to have your friends vote so that someone will respond: I registered a topic in 2014, and within a week it was the second highest voted issue for SQL Server. They are looking at my issue, but no promises on whether or when they might do something.

This will be interesting to watch.

One fundamental weakness of R is that it is slow. It is very high-level, has amazingly broad and deep libraries (primarily developed by all graduate level math/stats students) but it is "working" not optimized.

One of the core pieces of the RevolutionAnalytics value proposition was what it did to make R better. My low-level dinking with Rcpp gives 1000x speed improvements. I bet that Microsoft could do amazing things with R - becoming a primary source of open-source R.

I really wonder if MS has remembered that returning value to the customer is how to make/keep a market. Instead of approaching the customer as a revenue stream (think about how windows 8 and office 2013 are doing in the market) the approach of maximizing empowerment and delivery of new value is going to make/retain the market that Microsoft doesn't have right now. I hope they get a clue.

This has the potential to make R relevant to thousands or millions more people around the world.

I hope that this gives you the ability to execute on that potential. Congratulations.

@Mike:

-- I bet that Microsoft could do amazing things with R - becoming a primary source of open-source R.

Somehow I don't see the, by reputation, edgy personalities of R-Core cottoning to M$ in that way. After all, R has run on Windows for lots of years. How much has M$ contributed? That's a simple question, since I don't have any idea. There was nothing stopping M$ from contributing up until now, after all. Deepening the moat around Revo R; now that I can believe.

Question for the folks who are anti-MSFT.

Surely the beauty of the free and open source world is that if you don't like that a firm has been acquired then you can go elsewhere? Sure, you cannot take the commercial IP that Revolution Analytics created, but their contributions to the open source side are still there? So what's not to like?

I just see this as Microsoft realising that their analytic toolset is not strong enough for big data. Excel is a fine product but limited by row/column count. Microsoft implemented big grid to increase the rows/columns and also added HPC compute capability to link Excel to an HPC environment but I think they are just going for a mainstream approach to big data...

It would be really cool if I could use visual studio IDE in it's full glory to program R.

Also can you please replace DAX with R ;)
Am asking a bit there I know but worth a shot.

I like to cast a big vote for R in Visual studio! And not Excel. Excel is too constraining and my experience with Power Pivot and Power Query has been frustrating.

This is really exciting an interesting news. I am working on a research project for a FTSE100 in England and wondered who I would contact at MS about this as a possibility for our enterprise? It seems difficult to talk to anyone about possibilities with R?

The comments to this entry are closed.

Search Revolutions Blog




Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid
Get this blog via email with Blogtrottr