Distributed Computing Survey

[garo]

Hello All,
My wife and I are writing an academic article on why people participate in Distributed Computing projects. I posted a similar survey up here over a year and a half ago and got a large number of thoughtful responses. The findings were presented at a conference on Online Activism at the University of Southern California in 2003. The findings of this (final) survey will be written up for a forthcoming edition of the Journal for Computer Mediated Communication. We are a computer scientist (Anurag Garg) and a sociologist (Anne Holohan) respectively, and we appreciate very much you giving up a few minutes of your time to help us document the exciting, valuable and enjoyable experience of participating in distributed computing.
If you do not wish to post this information on the forum please feel free to send me a Private Message or email me at annie AT teamprimerib DOT com . The identity (including forum names) of all participants will be kept anonymous for the article. If you do not wish to answer any question (such as demographic Qs) please feel free to skip them.
Finally, for those who visit other forums, I am also posting this in Ars Technica and in the Free-DC forum.

Please answer each question in as much detail as you can. Thank you!

1. Are you
a. Male
b. Female

2. Are you between the ages of:
a. 0-18
b. 18-25
c. 25-50
d. 50+

3. Where do you live?
a. United States
b. Europe (specify which country)
c. Asia (specify which country)
d .South America
e. Canada
f. Australiasia including Pacific Islands
g. Africa

4. How did you get introduced to/first hear about Distributed Computing?

5. How many projects are you involved in?

6. Are you on any teams? If so, how many? Why did you choose that
particular one? Here please consider all teams from the same "organization" say Ars, or Anandtech or Free-DC as the same team.

7. Why do you give resources e.g. your computer power, your time - to the
project(s)? Can you describe exactly why this is rewarding for you?

8. How did you choose each particular project? What factors influence
this decision? (e.g. Project objective, Client suits your needs, Project Management, Server Reliability, Good Team etc.)

9. How computer literate are you? Do you consider yourself
a. An expert
b. Knowledgeable
c. A beginner?

10. Has distributed computing increased your knowledge of technology?

11. Do you know other people involved in the project (i.e. apart from online interaction)?

12. How frequently do you typically communicate with them?
a. Daily?
b. Weekly?
c. Monthly?
d. Every few months?

13. How many minutes or hours do you spend per visit to any Forum that
you participate in?

14. How would you describe your relationship with the other members of
a) your team b) the forum that team participates in? (Like work
colleagues/acquaintances/casual friends/close friends/strangers)

15. Do you trust the people you interact with in distributed computing and if so, why?

16. How would you feel if you discovered that somebody was not being
honest in their statistics? Will you report it, confront him/her or not worry too much about it?

17. Why is distributed computing important to you?

18. Do you think is distributed computing important to society and if so, why?

[Edit: Oops! Made a mistake with the email address . It is dot com and not dot net.]

[jebeagles]

1. a. Male

2. b. 18-25

3. a. United States

4.Friend of mine asked me to donate time to the UD Cancer project.

5. Two, GIMPS and UD Cancer project

6. no

7. I hate wasting things, and using extra CPU cycles to do research as opposed to letting them go to waste seems like a good idea.

8. Relatives have had cancer, so anything I can do to help find a cure, I'll do. And I love math, and I'm actually starting to get a grasp on some of the math behind GIMPS

9. b. Knowledgeable

10. Yes

11. A few friends of mine

12. a. Daily

13. about 1-2 minutes per visit

14. N/A

15. Yes, I guess I'm just a trusting person.

16. Not really worry about, because cheating on stats really doesn't make sense to me.

17. I get to help somebody do something, as opposed to sitting on the sidelines and watching as some great science and math is done.

18. It atleast gives computers an altruistic side, not the evil dark side shown in the media all the time whether via viruses or robots gone crazy, etc.

[ET_]

Please answer each question in as much detail as you can. Thank you!

1. a
2. c
3. b (Italy)
4. Reading an article about GIMPS in 1997
5. GIMPS, 321search, Operation Billion Digits, P-1
6. Yes, about 42 PC (But actually only 2 or 3 people). I choose this one because Italy needs some research push in Maths.
7. Any possible advance in Math research, that can be done with no loss of computing power, should be achieved.
8. I choose a project according to my possibility to really give a hand to it. (I was a former SETI enthousiast, but there was little 2reward" after years of computation, apart from the developing of complex computer algorithms).
9. a/2
10. Sure!
11. Yes
12. a
13. 90 minutes
14. (casual) friends)
15. Yes, because we all have the same target to reach.
16. I would not mind too much, only would warn him/her that some work may have been lost because of him/her.
17. Because you can reach a high ccomputational power with little effort.
18. It is from different points of view: it is ecologic, preventing misuse or peruse of electrical power; it is economical because the computational and economic power is shared; it is socially useful, because pushes people to join and have faith to one another.

Luigi Morelli

[edorajh]

1.
a. Male

2. Are you between the ages of:
c. 25-50

3. Where do you live?
b. Europe (Croatia)

4. How did you get introduced to/first hear about Distributed Computing? I read an article about SETI@Home in Astronomy Now

5. How many projects are you involved in?
One - GIMPS (Trial Factoring part only)

6. Are you on any teams? If so, how many? Why did you choose that
particular one? Here please consider all teams from the same "organization" say Ars, or Anandtech or Free-DC as the same team.
No

7. Why do you give resources e.g. your computer power, your time - to the
project(s)? Can you describe exactly why this is rewarding for you?
I like to track stats (rewarding is to get in Top 100 and more) and I like feeling of contributing to research goal with tousands of others members.


8. How did you choose each particular project? What factors influence
this decision? (e.g. Project objective, Client suits your needs, Project Management, Server Reliability, Good Team etc.)
Project objective, i.e. search for rare Mersenne primes is reason I choose GIMPS. I choose TF only segment of GIMPS because each assignement is completed for far less time than LL or DC.

9. How computer literate are you? Do you consider yourself
b. Knowledgeable

10. Has distributed computing increased your knowledge of technology?
Yes, a lot.

11. Do you know other people involved in the project (i.e. apart from online interaction)?
No

12. How frequently do you typically communicate with them?
-

13. How many minutes or hours do you spend per visit to any Forum that
you participate in?
30 minutes to 1 hour

14. How would you describe your relationship with the other members of
a) your team b) the forum that team participates in? (Like work
colleagues/acquaintances/casual friends/close friends/strangers)
casual friends

15. Do you trust the people you interact with in distributed computing and if so, why?
Yes. I feel we share similar goals and values.

16. How would you feel if you discovered that somebody was not being
honest in their statistics? Will you report it, confront him/her or not worry too much about it?
Not really worry about it. For me it's not very interesting to cheat on stats. The real deal is to achieve your stats by crunching. Then I can feel proud of them. Maybe I will report it to project management.

17. Why is distributed computing important to you?
It's fun to participate in project and track stats.

18. Do you think is distributed computing important to society and if so, why?
It is important as research tool.
Edo

[scottsaxman]

1. a. Male

2. c. 25-50, 29 years old

3. a. United States

4. I read about the SETI project in Discover magazine years ago. It's been all downhill since then!

5. Currently I'm involved in 3 projects. I have been involved in probably 12-15 at one time or another.

6. I am a member of Team-TNT.net. I choose them because they don't take things too seriously, are involved in many different projects, and allow members a great deal of freedom. They also know computers pretty darn well!

7. Contributing to things greater than yourself is very gratifying. It comes at no additional cost to me. Learning about different problems and people's attempts to solve them is a significant part of the process, also. I wouldn't be opposed to being involved in a project that I don't understand, as long as there were resources to learn about the project. My question is, why not DC?

8. I typically just need to match basic stuff like operating system, memory requirements, internet availability, etc., to try a project. The projects I have stuck with the longest, however, have tended to be smaller in terms of participants, have stable clients, and typically responsive managment. Good, frequently updated stats are also quite important. I don't need a pretty screensaver to run a project, though.

9. b. Knowledgeable

10. yes

11. Only a couple.

12. b. Weekly? - one is my father, the other is a good friend. I don't typically talk with them about DCing.

13. about 30 minutes a day, longer if I just started a new project

14. casual friends

15. I trust them. Tit for tat is my rule. They have given me no reason to distrust them.

16. I would be shocked, but would confront said person and expect him/her to resolve the issue, or else I'd report it.

17. DCing keeps me involved in areas of research that I have virtually no other way to connect with. And, as I said before, it's free and contributes to the greater good, whatever that may be.

18. DCing is important to society, though I expect society isn't largely aware of that yet. People doing research need processing power or bandwidth to keep science/math/etc. progressing, and they should have an opportunity to get it. DCing makes it cheap for them, easy for you, and makes a difference.

[garo]

Thanks guys, your replies are very useful. PLease keep them coming.

[mephisto]

1.
a. Male

2. Are you between the ages of:
c. 25-50

3. Where do you live?
b. Europe: Norway

4. How did you get introduced to/first hear about Distributed Computing?
Article online somewhere, some computing or science publication around 2001, I think.

5. How many projects are you involved in?
One. (One DC, but currently thinking about getting involved in an open source programming project; presumably Squeak - an open-source implementation of Smalltalk)

6. Are you on any teams?
No.

7. Why do you give resources e.g. your computer power, your time - to the
project(s)? Can you describe exactly why this is rewarding for you?
I think it's a variation on the 'being part of something greater than yourself' theme. A whole string of random incidents may decide whether I run DC, contribute to an open source project like Linux, or distribute food and clothing to the poor.
But there are certain 'necessary but not sufficient' criteria for any project I would consider contributing voluntarily for: I must find it iteresting (at least for the time being), it must be non-commercial, it must have at least some kind of potential real-world value (for me, that excludes SETI, as I don't believe in ET) etc.

8. How did you choose each particular project? What factors influence
this decision? (e.g. Project objective, Client suits your needs, Project Management, Server Reliability, Good Team etc.)
Various quite (but not completely) random factors. I got interested in DC in general because I think it's a neat idea technically, I started with Prime95 because I saw a list of DC projects where this was the one with the largest potential monetary reward, I stayed beyond the first LL because I found quite a lot of iteresting information while browsing the forum.

9. How computer literate are you? Do you consider yourself
a. An expert
Depends. 'Quite knowledgeable' the label I would use. Master in computer science/artificial intelligence, programmer for 4 years. Approaching 'expert' status in certain areas, but not hardware, networking, mainframe, assembler or Win API expert, for example.

10. Has distributed computing increased your knowledge of technology?
Yes. Mostly indirectly through links in this forum. Examples: P4 architecture, signal processing, Fourier transforms

11. Do you know other people involved in the project (i.e. apart from online interaction)?
No.

12. How frequently do you typically communicate with them?
-

13. How many minutes or hours do you spend per visit to any Forum that
you participate in?
Per visit? Couple of minutes average.

14. How would you describe your relationship with the other members of
a) your team b) the forum that team participates in? (Like work
colleagues/acquaintances/casual friends/close friends/strangers)
-

15. Do you trust the people you interact with in distributed computing and if so, why?
Trust? I don't think Prime95 is a virus, or that everybody on the forum is trying to steal my money, it that's what you mean. I probably wouldn't lend them 10 grand either.
I'm merely computing, beyond a fairly basic level, trust doesn't enter the picture.

16. How would you feel if you discovered that somebody was not being
honest in their statistics? Will you report it, confront him/her or not worry too much about it?
Confront or not worry, depending on my level on engagement there and then.

17. Why is distributed computing important to you?
It's not. Not in particular. It's one of several options in a more general class of 'what I might want to do with my spare time'.
'Being part of something larger than myself', 'potentially contributing to society', 'learning stuff' etc. are factors that are not unique to DC - see question 7.
See also Eric Raymond (well-known open source advocate): Homesteading the Noosphere
( http://www.catb.org/~esr/writings/ho.../homesteading/ )
for his theories on what drives the open source community.
By and large I agree with Raymond, and at least in my case, it explains to a large extent my interest in open source as well as my contribution to DC.

18. Do you think is distributed computing important to society and if so, why?
I think it has the potential to be. Both in the technical sense (including large clusters of cheap computers for scientific and business use) and in the 'social structure' sense (lots of people voluntarily spend time, effort and computer resources on a computing problem of general interest).
I don't think current DC (social sense) is important to society: I cannot readily see that society would be much worse off without SETI or knowledge of certain huge primes.
But I do think current DC is just the frontier, and current participants are akin to frontiersmen (and -women :). More so in a project like prime95, which requires aboove-average dedication, than in a project like SETI.
DC is still in a starting phase. I think DC will grow on us, and I don't find it impossible that DC at some point will be as ubiquitous as internet is today. At that point, it will be important to society.


BTW: Is your paper from the previous survey available on the net somewhere?

[mephisto]

Addition to 8 (sometimes I *really* dislike the edit time limit )

- project objective must be something I find somewhat interesting and with at least some degree of social or scientific value. Contrast with contributing to open source projects, which is far more work, and which I would only consider if the project objective is *very* interesting or something I actually need.
- client, project management, server reliability and so forth are only important to the extent they don't work. That is, I won't choose a project because of a reliable server (or stable management, or..), but I might leave a project with an unreliable one. Of course, if the client doesn't run on my platforms, I won't enter in the first place.

[garo]

mephisto,
Not yet, but I can make it available to you privately - it was a presentation not a paper. The current survey should result in a published paper which will be available.

[patrik]

1. Are you
a. Male

2. Are you between the ages of:
c. 25-50

3. Where do you live?
b. Europe (specify which country) Sweden

4. How did you get introduced to/first hear about Distributed Computing?
At mersenne.org. I don't remember where I heard about mersenne.org.

5. How many projects are you involved in?
One. Or two, if manually selecting numbers to run ECM curves on also counts as a project different than GIMPS.

6. Are you on any teams? If so, how many? Why did you choose that
particular one? Here please consider all teams from the same "organization" say Ars, or Anandtech or Free-DC as the same team.
No.

7. Why do you give resources e.g. your computer power, your time - to the
project(s)? Can you describe exactly why this is rewarding for you?
I like to do the math (eliminating prime candidtates and finding factors). I would have been doing this anyway, if I had had the time, even if there were no coordination and project. Now I can do it without spending so much time.

8. How did you choose each particular project? What factors influence
this decision? (e.g. Project objective, Client suits your needs, Project Management, Server Reliability, Good Team etc.)
I don't. GIMPS just happened to be the first one I came across and liked. There are still projects I might join (like 17-or-bust), but I haven't had time to read about them yet. But there are projects I don't like, e.g. where you just get some data to process and have no idea yourself exactly what your computer are doing right now.

9. How computer literate are you? Do you consider yourself
b. Knowledgeable

10. Has distributed computing increased your knowledge of technology?
It has increased it about computer hardware.

11. Do you know other people involved in the project (i.e. apart from online interaction)?
No.

12. How frequently do you typically communicate with them?
n/a

13. How many minutes or hours do you spend per visit to any Forum that
you participate in?
No idea. A wild guess is three hours on average.

14. How would you describe your relationship with the other members of
a) your team b) the forum that team participates in? (Like work
colleagues/acquaintances/casual friends/close friends/strangers)
Well, I'm not in a team so strictly speaking the question doesn't apply to me, but speaking about the forum for the projects, the best way to describe it is like members of a club for practicing a hobby, like collecting coins etc. This is a hobby for me, I think.

15. Do you trust the people you interact with in distributed computing and if so, why?
I guess so. Most of the time they seem to get the facts right.

16. How would you feel if you discovered that somebody was not being
honest in their statistics? Will you report it, confront him/her or not worry too much about it?
I wouldn't worry too much about it. Unless that person was also sending fake results.

17. Why is distributed computing important to you?
It is important in the way it prevents duplicating work. And if you include the discussion forum in the concept of DC, that is also important, since it is a good way to meet others with similar interests.

18. Do you think is distributed computing important to society and if so, why?
Not really. But it's fun.

Feel free to copy from my answer last survey, http://www.mersenneforum.org/showpos...55&postcount=6. I didn't say it so well this time.

[dswanson]

1. Are you
a. Male

2. Are you between the ages of:
c. 25-50

3. Where do you live?
a. United States

4. How did you get introduced to/first hear about Distributed Computing?
Newspaper article - San Jose Mercury News, following discovery of M37 (3021377) in 1997.

5. How many projects are you involved in?
One

6. Are you on any teams? If so, how many? Why did you choose that
particular one? Here please consider all teams from the same "organization" say Ars, or Anandtech or Free-DC as the same team.
Define "team". Strictly speaking, no, I'm not on a team. All of my stats are derived from machines I run myself. But for nearly two years one other participant and I have been closely coordinating a massive doublechecking project, so indirectly you might consider us as a team. We just keep our stats separated.

7. Why do you give resources e.g. your computer power, your time - to the
project(s)? Can you describe exactly why this is rewarding for you?
Thrill of the hunt. I don't really expect to find a new prime, particularly since I concentrate my efforts on doublechecking. But it's fun to watch the progress of the project as a whole, and to know that I'm making a significant contribution to that progress.

8. How did you choose each particular project? What factors influence
this decision? (e.g. Project objective, Client suits your needs, Project Management, Server Reliability, Good Team etc.)
GIMPS was the first project I tried, and it's the one I remain with. I started with it about 6 months before SETI@Home came out, thinking I'd stick with GIMPS until the SETI client was available. I switched to SETI for a few months, but came back to GIMPS because (1) I didn't care for the screensaver that I couldn't turn off, (2) SETI quickly picked up so many participants that I didn't feel my contribution mattered much, and (3) I liked the openness, volume, and update frequency of the GIMPS statistics posted on Primenet. Responsiveness of project management (ie, George) has also been a big plus for GIMPS.

9. How computer literate are you? Do you consider yourself
b. Knowledgeable

10. Has distributed computing increased your knowledge of technology?
Yes. I'd never have tried to build a farm of headless computers without it. That was a real learning experiences: Linux, hardware, networking, firewalls, file servers, and more!

11. Do you know other people involved in the project (i.e. apart from online interaction)?
None offline.

12. How frequently do you typically communicate with them?
-

13. How many minutes or hours do you spend per visit to any Forum that
you participate in?
30 - 60 minutes

14. How would you describe your relationship with the other members of
a) your team b) the forum that team participates in? (Like work
colleagues/acquaintances/casual friends/close friends/strangers)
For my loosely-defined team, I'd say casual friends. For the forum, I'd say acquaintances.

15. Do you trust the people you interact with in distributed computing and if so, why?
Yes. Nobody's ever given me any reason not to. This is a great bunch of people to interact with.

16. How would you feel if you discovered that somebody was not being
honest in their statistics? Will you report it, confront him/her or not worry too much about it?
I'd be disappointed in that person, but overall I wouldn't worry about it too much.

17. Why is distributed computing important to you?
It's a hobby I enjoy greatly.

18. Do you think is distributed computing important to society and if so, why?
Yes, although the reason varies depending on the project. Some of the commercial-oriented projects may have near-term monetary or societal impacts. Some of the more academic ones (such as GIMPS) won't have any immediate impact, but the history of science shows that things pursued for no particular reason other than curiousity often end up having practical benefits many years later. And some projects, such as SETI@Home, are unlikely to succeed at all. But if it does, it will have an enormous impact in how the human race sees itself and its place in the universe.