In my previous company I posted the same idea on their worldwide bulletin board (Yammer.com) and sadly didn't get any official reply. It can be about operating costs and hurdles, but I think most of all about setup costs and hurdles vs. expected PR benefits for the company.

I think the crucial step would be to make WCG well enough known in the media so that companies would seek by themselves to contribute in order to get good PR. It's all about making yourself viral.

I can also tell you about my difficulties when trying to install this at work on my own. A common problem is that even if the PC is able to run Boinc, its fan makes too much noise. Even if you cap CPU time as low as 20% or so the problem may persist, or even become more annoying because Boinc modulates CPU time over a long time -- e.g. the fan makes a lot of noise for 1 second, remains silent 4 seconds, and repeats this cycle over and over.

Two employers ago I had this problem. Eventually I kept Boinc installed and got rid of the excessive fan speed by relocating the chassis in a weird way (laying down on the floor 90 degrees instead of standing, and resting on four rubber erasers as table feet). This machine kept churning two years after I left the company, I guess until someone reformatted and reinstalled Windows.

At the next company I didn't have admin access to my own PC. As I said I raised the initiative informally but got no response.

At my current company I have a Windows laptop and a Linux workstation. I tried the laptop and although it's much more powerful than my home PC, the fan makes very noticeable noise even at 20 or 10%. I disabled it before any coworker would complain about the noise. The Linux distro on the workstation (OpenSuse) does not provide a working Boinc version from its repository. I did make it work after a lot of work and requiring a lot of help from the technical forum here, but not yet in a nice automatic way. Moreover these Linux machines of ours are already used for background distributed computing (distributed compilation of our own code), so in the end it looked to me that I should not leave it running.

You can imagine how these difficulties (even forgetting about my last case) could play out catastrophically if you mass install Boinc across a whole organization.

So Boinc could also help this by being more foolproof and robust, and allowing more customization by the sysadmin and requiring less control from the user.

I used to work for a MAJOR corporation in one of the top 5 most populous cities in the US. At the time this company had a little over 10,000 employees and most employees had laptops and desktops so you imagine the number of machines. This doesn't include anything in the two data centers each with multiple computational floors. I scheduled a lunch with the CIO and made an informal presentation to run BOINC. The items I mentioned above were just a few of the point that were raised by the CIO. I knew going in I wasn't going to get approval on a money basis as there weren't any cost savings to be had anywhere. I approached it more from a give back to the community stand point and what a good thing it would be if a discovery happened from the research and the company was identified as having contributed to that discovery. In other words, all companies like to look good in the eyes of their customers. The CIO didn't think the benefit was worth the potential cost to the IT budget in light of the fact that IT was always having to justify their budget by showing a business benefit. The major corporations are older, stodgy, entities and they aren't concerned about "ethics" unless it's business ethics forced on them by regulators. They are only concerned about earnings and shareholder value. If it doesn't contribute to those two things, it's a tough sell. It didn't work at my company but my recommendation would be to pursue it through Corporate Communications or Community Outreach initiatives.

It wouldn't have to be a default app but could be a downloadable app from a corporate menu of apps. The employees could choose to contribute but the support issue is still there.

I should also mention that there were a group of us that had it installed and it was working fine for several years and then the infamous IBM decided to move some stuff to the CDN. That pretty much killed it at that corporation. We were accessing IBM through a HTTPS proxy and we were able to get it working with few tweaks and corporate security and network support were very helpful but the CDN thing was more than they were willing to support. The CDN doesn't use the same IP address all the time and folks behind firewalls were having issues also. The recommendation from IBM was to unblock a RANGE of IP addresses and they provided the range. Our corporate security folks wouldn't not support that at all so it died as things moved to the CDN.

Focusing on big companies, there are roughly 1,900 companies in the U.S. with over 5,000 employees. http://www.census.gov/epcd/susb/latest/us/US--.HTM

To think about the potential for a moment: 1900 x 5,000 = 9.5 million (people). Not all of those employees are sitting at a computer, so let's round down to 9 million office computers, in large companies, as the target.

10% of 9m is 900,000 machines. 5% of 9m is 450,000 machines - a massive amount of processing power.

@Doneske, as a person who works in corporate America, I definitely believe you when you recount your experiences. There is no doubt that there would need to be buy-in by the company leadership.

Perhaps the model could be along the lines of this:
Approved by management
- They are aware of it, they allow it through firewalls, etc.
Grassroots maintained
- Management is allowing it, and taking credit for it! It is a "program" that they can trumpet. But the employees are doing the installation.
A cap on how many employees are allowed to do it.
- Perhaps a set number per team or something like that.

Obviously, this model was not followed by the company you were working at, @Doneske. But if it were to exist, it could then be pointed at by employees who want to make this happen. It would be "a thing". That's not to say that your company would suddenly say, "Ok, well then we'll do it." But at least it would be out there.

The single most important thing these days is social media. Get on Facebook, Twitter, sing the praises of this work.

I support WCG on my sciencesprings blog. Every time I see a good article I do a blog post.