Suggested Changes to F@h Website

[Jesse_V]

Hi, there was an experienced Wikipedian who just called me on this one, so I feel that I need to get this statement straight.
At http://folding.stanford.edu/English/FAQ-main#ntoc4 it says:
"Unlike other distributed computing projects, Folding@home is run by an academic institution (specifically the Pande Group, at Stanford University's - Chemistry Department), which is a nonprofit institution dedicated to science research and education. We will not sell the data or make any money off of it. Moreover, we will make the data available for others to use. In particular, the results from Folding@home will be made available on several levels. Most importantly, analysis of the simulations will be submitted to scientific journals for publication, and these journal articles will be posted on the web page after publication. Next, after publication of these scientific articles that analyze the data, the raw data of the folding runs will be available for everyone, including other researchers, here on this web site."

There's a couple of statements here that I need to confirm because they seem almost contradictory. First, not all the papers can be accessed for free. While I've since learned that some journals can have the publication under a paid subscription, that seems a little deceptive to me in light of this statement. And where is the raw data runs? I know I can get to the data for the 1.5-millisecond record-breaking simulation, but I haven't been able to locate any others. If indeed they do exist there should be a link on that FAQ and in other areas of the website.

Like I said I was called on it, and then I started agreeing that there was indeed a problem. Perhaps you can shed some light on what is going on here. I just wanted to make sure the website is up-to-date and accurate, and since this statement shouldn't be changed/removed, I just wanted to see some backup. Thanks.

[VijayPande]

A few comments:

* "not all the papers can be accessed for free". Due to new NIH rules, all papers have to be free, but *after* a year. This is the way science works and I unfortunately can't change how peer reviewed journals work single handedly. But, NIH's power has led to this compromise, which I think is reasonable (journals get something, but the papers are eventually out there to everyone without a fee). Also, note that essentially all scientists (at Universities and companies) have free access to these journals, due to institutional subscriptions. Finally, note that we (the scientists, member of the FAH team, etc) don't see any money when the journals sell the papers (don't get me started about that one, but again it's how science works and not something I can change).

* "And where is the raw data runs?" We have been making these available in general on request and in cases where people ask for data sets repeatedly (eg https://simtk.org/home/foldvillin), we make them available on a website linked from folding.stanford.edu. The big problem is hosting a web site with terabytes of data – it makes sense only to do that in cases where there is sufficient demand that we can't handle it individually.

[Jesse_V]

A few comments:

* "not all the papers can be accessed for free". Due to new NIH rules, all papers have to be free, but *after* a year. This is the way science works and I unfortunately can't change how peer reviewed journals work single handedly. But, NIH's power has led to this compromise, which I think is reasonable (journals get something, but the papers are eventually out there to everyone without a fee). Also, note that essentially all scientists (at Universities and companies) have free access to these journals, due to institutional subscriptions. Finally, note that we (the scientists, member of the FAH team, etc) don't see any money when the journals sell the papers (don't get me started about that one, but again it's how science works and not something I can change).

* "And where is the raw data runs?" We have been making these available in general on request and in cases where people ask for data sets repeatedly (eg https://simtk.org/home/foldvillin), we make them available on a website linked from folding.stanford.edu. The big problem is hosting a web site with terabytes of data – it makes sense only to do that in cases where there is sufficient demand that we can't handle it individually.

Once again, I'm honored by your response and thank you very much for clarifying. The answer to the papers question is an interesting compromise, and the information about the data runs makes sense. I'm glad it backs up what's on the website and now that you've described what you have to balance I'd have to agree that seems the best solution.

Again, thank you for explaining.

[Jesse_V]

I have another issue. There's a Wikipedian who is calling into question the statement on the F@h Wikipedia article that says that Folding@home has produced more papers than "all other major distributed computing projects combined". This statement is backed up by the F@h website on this page: http://folding.stanford.edu/English/FAQ-main#ntoc7, but I'm now agreeing that that statement is no longer accurate. In our conversation here: http://en.wikipedia.org/wiki/Talk:Folding@home#Lead he gathered information from various BOINC websites, and counted all the papers produced from what he considered "major" distributed computing projects. The total was 173, and F@h has produced 95. He defined major as "if it has over 10,000 active volunteer members or currently sustains over 250 x86 TFLOPS" which seems reasonable to me, but "major" is not really defined anywhere else. So, I believe the statement should be removed or modified. I want to make sure the F@h website stays updated, because it wouldn't speak well for us if it wasn't. He suggested something to the effect of "more than any other individual distributed computing project" but noted that that statement would have to be verified. So if anyone could make the change or suggest something different, I'd appreciate it.

[VijayPande]

that text on our web site is pretty old so I can imagine that times have changed. good point that we should revise it. done!

[Zagen30]

I noticed a few things that could use some updating. First off is the minimum requirements section of the main FAQ:

All computers can contribute to Folding@home. However, if the computer is too slow (e.g. wasn't bought in the last 3-4 years or so), the computer might not be fast enough to make the deadlines of typical work units. A Pentium 3 450 MHz or newer equivalent computer (with SSE) is able to complete work units before they expire.

Now, as far as I know a P3 is still capable of meeting the uniprocessor deadlines, but they're very old at this point, certainly much more than 3-4 year old technology. Some clarification would be useful to potential members who recognize this. There's also no mention of GPUs at all; evidently this was written before even GPU1, but it might not be a bad idea to put in the requirements there, as while that information is listed under the GPU-specific FAQs, a first-time visitor may not have read those yet. It could also tip them off to the existence of GPU folding, which they may not recognize; I noticed there's no mention of GPUs at all in the main FAQ.

Second, under the GPU3 FAQ, the section titled "The third generation GPU core (GPU3) for ATI (March 2011)" just has the placeholder "OpenCL, OpenMM, fahcore_16, V7, etc." Presumably someone meant to get back to it and never did. I recognize that with v7 hopefully getting a full release in the next couple of months it may not be all that necessary to fix this.

I also noticed the link to the SHARPEN FAQ is broken. That's probably not PG's fault, but having broken links still doesn't look good.

Finally, does the main FAQ still need the section on the screensaver? That hasn't been a feature for a couple of client versions now. I realize that it's at the bottom of the FAQ and is thus somewhat unlikely to be read, but I suppose it could give someone the wrong impression.

[7im]

The typical work unit is SMP and GPU, so yes, your computer needs to be recent. However, the min spec to complete a CPU fahcore 78 work unit is still the P3 listed, although not so typical any more. I suppose one could make it more clear the P3 is the min spec, while the recent spec is the recommended spec.

As soon as the Min Versions for all servers on the PSummary page are 5.0 or above, then we can remove the Screen Saver information.

I'll add GPU right below the mention of SMP... although GPU has several links from the FAQ directory page, so it's probably not getting missed.

[Jesse_V]

As soon as the Min Versions for all servers on the PSummary page are 5.0 or above, then we can remove the Screen Saver information.

I don't understand why this is a requirement. Yes, if all the servers supported 5.0 or above I imagine that anyone (is there anyone?) who is still running screensavers couldn't get work. F@h clients can't be mirrored, so the only place to get them is on the F@h site. Are they even still available? I have not been able to find them. Removing that information from the FAQ won't hurt those people, because if the screensaver client can't be downloaded off of the website, its obviously been working fine for them for years. If they run into issues, they can post on the forums and we can tell them to upgrade to v7. The information about the screensavers does more harm than good. It implies that something exists when it very likely no longer does thus confusing any newcomers, but caters to a tiny number of people, if anyone at all. Their screensavers will continue to work no matter if the information is on the FAQ or not.

[codysluder]

I noticed a few things that could use some updating. First off is the minimum requirements section of the main FAQ:

All computers can contribute to Folding@home. However, if the computer is too slow (e.g. wasn't bought in the last 3-4 years or so), the computer might not be fast enough to make the deadlines of typical work units. A Pentium 3 450 MHz or newer equivalent computer (with SSE) is able to complete work units before they expire.

At the very least, add say something similar to "...provided it runs 24 x 7 whereas more modern computers tend to be much faster so they can meet the deadlines of the longest even if they don't run 24x7."

That may be a good place to mention the need for speed and it's relationship to points, too.

[Jesse_V]

That may be a good place to mention the need for speed and it's relationship to points, too.

Points represent scientific accomplishment. Since the latter is clearer, I recommend saying "for better scientific productivity" rather than "for more points". There is a great scientific need for the rapid completions of Work Units, as the overall simulations take less time to complete.

[7im]

As soon as the Min Versions for all servers on the PSummary page are 5.0 or above, then we can remove the Screen Saver information.

I don't understand why this is a requirement. Yes, if all the servers supported 5.0 or above I imagine that anyone (is there anyone?) who is still running screensavers couldn't get work. F@h clients can't be mirrored, so the only place to get them is on the F@h site. Are they even still available? I have not been able to find them. Removing that information from the FAQ won't hurt those people, because if the screensaver client can't be downloaded off of the website, its obviously been working fine for them for years. If they run into issues, they can post on the forums and we can tell them to upgrade to v7. The information about the screensavers does more harm than good. It implies that something exists when it very likely no longer does thus confusing any newcomers, but caters to a tiny number of people, if anyone at all. Their screensavers will continue to work no matter if the information is on the FAQ or not.

Pande Group has always taken an inclusive position. For example, continuing to provide work units for PPC Macs, and Pentium 3s as long as possible as not to exclude those machines from being able to contribute to the project.

For example, there are more than 1000 active clients still running Windows 2000, which does not support v6 or V7. And ~65,000 people still running Windows XP, a 10 year old OS (Oct 25th, 2001). Clients installed in the last 10 years could still be running v4, although I have no stats on v4 usage.

But with Windows 7 actually getting traction in the market (user acceptance, installed numbers surpassing XP), I expect a lot of XP users will finally upgrade, while the Windows 2000 server people probably won't if they haven't updated already.


Added: Does Folding@home run on my graphics chip or GPU?

Added: Does Folding@home run on my Sony Playstation 3 game system (PS3)?

[codysluder]

I expect a lot of XP users will finally upgrade

I see no good reason to do that. If it's working, it's a waste of money. When the computer dies, it will be replaced with one that has a new OS on it.

I can see absolutely no difference in how well FAH runs on XP or on later Microsoft versions.

[7im]

As soon as the Min Versions for all servers on the PSummary page are 5.0 or above, then we can remove the Screen Saver information.

I don't understand why this is a requirement. Yes, if all the servers supported 5.0 or above I imagine that anyone (is there anyone?) who is still running screensavers couldn't get work. F@h clients can't be mirrored, so the only place to get them is on the F@h site. Are they even still available? I have not been able to find them....

Yes, still available. See the Older Clients link from the Download page. For example, I forgot how simple v4 (built Dec. 8th, 2003) was to install and run...

Code: Select all

--- Opening Log file [November 8 18:37:10]
# Windows Graphical Edition ###################################################
###############################################################################
                       Folding@home Client Version 4.00
                          http://folding.stanford.edu
###############################################################################
###############################################################################

[18:37:39] - Ask before connecting: No
[18:37:39] - User name: 7im (Team 14)
[18:37:39] - User ID = 4794388B3CB2D04A
[18:37:39] - Machine ID: 1
[18:37:39]
[18:37:39] Work directory not found. Creating...
[18:37:39] Could not open work queue, generating new queue...
[18:37:39] Initialization complete
[18:37:39] + Benchmarking ...
[18:37:42] - Preparing to get new work unit...
[18:37:42] + Attempting to get work packet
[18:37:42] - Connecting to assignment server
[18:37:42] - Successful: assigned to (0.0.0.0).
[18:37:42] + News From Folding@Home: Welcome to Folding@Home
[18:37:42] Work Unit has an invalid address.
[18:37:42] - Error: Attempt #1  to get work failed, and no other work to do.
             Waiting before retry.
[18:37:54] + Attempting to get work packet
[18:37:54] - Connecting to assignment server
[18:37:54] - Successful: assigned to (0.0.0.0).
[18:37:54] + News From Folding@Home: Welcome to Folding@Home
[18:37:54] Work Unit has an invalid address.
[18:37:54] - Error: Attempt #2  to get work failed, and no other work to do.
             Waiting before retry.

If I don't get a WU by tomorrow, the v4 stuff will come off the FAQs.

[7im]

I expect a lot of XP users will finally upgrade

I see no good reason to do that. If it's working, it's a waste of money. When the computer dies, it will be replaced with one that has a new OS on it.

I can see absolutely no difference in how well FAH runs on XP or on later Microsoft versions.

Agreed. However, not all systems are dedicated to fah 24/7. Some people actually like to use their computers for other things, and they tend upgrade once in a while. Not my recommendation, but it happens.

For a dedicated box, a quicky update of ubuntu once in a while works well too.

[Jesse_V]

Found some other errors. On this page: http://folding.stanford.edu/English/About there are a couple of deadlinks that hopefully can be taken care of. I'm not sure exactly what they should point to, only that they don't currently work as expected. There are two of them, and both links are the first several words of two paragraphs:
1) The bit about Erik Lindahl's group: http://folding.stanford.edu/English/About#ntoc8
2) The short description of Bojan Zagrovic's lab: http://folding.stanford.edu/English/About#ntoc11

Thanks.