The Dilution of Distributed computing

narzy

Elite Member
Feb 26, 2000
7,006
1
81
First off a question, I've had to uninstall BONIC it is causing hell with VS.Net and some other strange programs I use on a daily basis, it tends to make my system a bit unstable and sluggish in games. (no I'm not OCing and my processor temp is fine...) so I'm looking for another project outside of seti to be a part of that doesn't use BONIC. (plus adding machines with the bonic user key is a pain in the arse.)

now for the news,

I've started my own web hosting company and I wanted to extend an offer to you guys for inexpensive hosting, the company is TechMayhem.net I've set up a special account for my DC bro's here are the features.

For $2.95 per month
Features:
150MB Storage
4500MB Data-Transfer per month
Cpanel

your account has unlimited:
E-mail addresses
E-mail forwarders
E-mail Lists
Domains
Subdomains
Parked Domains
MySQL Databases
FTP accounts

If your interested send me a PM or E-mail and I'll get you all set up. here is the direct link to the payment processor (I use paypal right now) payment link

Now for the commentary, This is a peice I am writing for my review website, but I thought I would get you guys to give it a once over and get some feedback.

DRAFT The dilution of Distributed computing DRAFT

Ever since Seti@home revolutionized the way we process complex tasks by dividing the problem into smaller problems and having home computers solve those smaller problems to create and end result of the big picture distributed computing has become a hit amongst home computer enthusiasts that have an interest in the sciences. However because of the success of DC a multitude of new projects have materialized. This in general is a good thing revolutionizing, spawning new projects and allowing the diversification of Distributed computing projects as a whole. Unfortunately there are unintended consequences to this evolution and that is the watering down of resources available to any given project.

When distributed computing began to take shape there were only a few projects to choose from, these included Prime95, RC5-64, and Seti@Home. During that period of time the resources that groups were able to put in to each project were immense creating teams of people working together under a common name, this created rivalry between teams from good old fasion team ribbing to some very emotional times between teams. A careful game of politics had to be played not only inner-team but also with other competing teams, This rivalry boosted moral and distributed computing participation as groups were at it to "defend their honor."

Today the scene is much different then it was 4 years ago, there are more projects available, which for all intensive purposes is a good thing however it has led to the slow dilution of massive power available to any given project at any given time. The rivalries have died down if not to have vanished completely and the organization that once existed has been shattered and has splintered off into a myriad of different directions. There is no one to "blame" for this nor is it really a "problem" but more of a realization of the state that DC is in and the evolutions that have occurred throughout the involved community. Today the feel of the DC community is much different then it was with so many dynamic people involved in so many projects its becoming increasingly challenging for communities to hold together the bonds that not long ago seemed inseparable.

Unfortunately there also seems to be another problem plaguing the DC community, Recruitment. Getting people involved has become extremely challenging. There are a number of uphill battles that the community has to fight to get people interested and to get them to participate. A strong detractor from DC is the "myth" that because your using your computer at "100%" that it will shorten the life of your computer or the extra heat produced is damaging, people make the comparison to a car engine while not understanding the distinct differences between mechanical and electrical components. They believe to be "productive" they have to leave their computer on 24/7 and that its helpless to contribute if they don't. These are only a few of the numerous issues that have to be faced to get people to participate while also trying to defeat the "lack of interest."

The cause of the complete lack of interest in the DC projects boils down to results, while many of us today are happy with our stats and watching those numbers climb is gratifying enough to keep working on the project the average Joe wants to see results and wants to see them now! There are inherent problems to this way of thought, it is valid. on the topic of results interpretation of the results given to the general DC Joe can be hard if not impossible to understand unless your imamate familiar with the project and sciences behind it, if people are to participate in a hobby such as DC they don't want to have to get a doctorate just to understand what they are doing.

Key points the DC community needs to address:

? Dissolving the Myths
? Show constant and consistent progress
? Show progress in ways the general population can understand without having to know a lot of background information on the subject(s) at hand while still providing "meaningful" data
? Introduce a more interactive component to DC, allow "set it and forget it" but also allow people to become actively involved if they choose to.

 

IEC

Elite Member
Super Moderator
Jun 10, 2004
14,600
6,084
136
Hmm... well DC projects in general HAVE been diluted, mostly due to 1) a lot more DC projects 2) less recruitment 3) people care less :(
 

BlackMountainCow

Diamond Member
May 28, 2003
5,759
0
0
Indeed, whenever I ask ppl to do a DC, the sentence "that will kill my PC earlier" has almost become a standard answer. :(
 

narzy

Elite Member
Feb 26, 2000
7,006
1
81
also this is a work in progress so I am updating it right now, I'll post again when I'm done updating so you know you can read the "entire" story.
 

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,695
4,657
75
Hi, Narzy! Long time no see!

YGPM.

It's true the rivalries seem weaker on this team. But it seems like it's more from a general inability to keep up in all projects on our part. Committed projects like SETI still have strong rivalries. I wonder whether other teams that have dedicated project forums, like Free-DC, are also weakening this way.
 

BlackMountainCow

Diamond Member
May 28, 2003
5,759
0
0
Originally posted by: Ken_g6
I wonder whether other teams that have dedicated project forums, like Free-DC, are also weakening this way.

Wow, that's a very good question. I didn't even know that other forums have that!

 

narzy

Elite Member
Feb 26, 2000
7,006
1
81
thanks for the corrections, I'm dumping it in to word now to clean it up but I fear this version is going to have to stick around a bit mainly because I don't want to "redo" fusetalks formatting crap.

This article can also be tuned to the OSS community with a few modifications.
 

narzy

Elite Member
Feb 26, 2000
7,006
1
81
wow I'm kind of suprised more people haven't chimed in on this.
 

Insidious

Diamond Member
Oct 25, 2001
7,649
0
0
ding dong (my little chimes :D )

Things that make it more difficult for me to stay engaged:

  • lack of results I can understand (or none at all)
  • Condescending attitudes of those on the project "home forum" to anyone outside their clique
  • Difficulty convincing friends to join in with me

Things that make it difficult to get friends to join in:

  • Potential users are afraid of intrusion into their private space by allowing use of their PC by strangers (afraid of getting "hacked")
  • No way to know what is really being computed on the machine by the client
  • Fear of damaging computer with 24/7/100% usage

These are legitimate concerns. Particularly the one about having to trust that what is being said to be computed is really what is being done. The sad lack of tangable results doesn't help reduce the mystery of what is really being computed. For all I know Uncle Sam might be using my computer to spy on other computers... or me!

-Sid
 

BlackMountainCow

Diamond Member
May 28, 2003
5,759
0
0
What Sid wrote is very much the truth. There are only a few projects that have a close feedback to the crunchers, like SETI with their focused re-searches of hotspots in the sky or FaD with their anti-cancer /anti-aids tests & certificates.
 

bluestrobe

Platinum Member
Aug 15, 2004
2,033
1
0
It's society in general.

People are less prone to volunteering and expecting results without effort. I think lack of interest in DC projects is part of the bigger picture in the world today. DC projects cost people to run on their computers, I only have 7 systems running it and I notice it on my power bill. Most people won't touch anything unless there?s something in it for them or they can see immediate results. Selfish Greed.
 

n0cmonkey

Elite Member
Jun 10, 2001
42,936
1
0
Nice article narzy. :beer:

Nice deal too. If a friend of mind didn't beat your price by $2.95, I'd be jumping on it. ;)
 

Spacehead

Lifer
Jun 2, 2002
13,067
9,858
136
Nice offer on the web hosting.

? Dissolving the Myths
Part of the problem about dissolving myths about DC is that most people just plain don't know much about their computers in general to begin with. Alot of people don't know enough to blow the dust out of computers every so often even if they aren't running DC.
Security is also a big issue i get when i've asked about running DC project. It's a legitamit question to ask but these are the same people that would(& have) dowload & run some spyware infested program without blinking an eye.

? Show constant and consistent progress
? Show progress in ways the general population can understand without having to know a lot of background information on the subject(s) at hand while still providing "meaningful" data
That's kinda tough. Even if the info's there, people just want the end results. Example- they don't care how cancer was cured, just that there's a cure for it.

? Introduce a more interactive component to DC, allow "set it and forget it" but also allow people to become actively involved if they choose to.
I know Dave's said this for years- "...convince Microsoft to include DC as a drop down Menu option."
I wonder if BOINC will be a first step in that direction.
 

Ken g6

Programming Moderator, Elite Member
Moderator
Dec 11, 1999
16,695
4,657
75
Originally posted by: dmcowen674
Best bet like I've said from the beginning is to convince Microsoft to include DC as a drop down Menu option.

Until that happens, DC will be like CB Radios.
Whatever happened to the thing Google was putting in their toolbar? F@H, wasn't it?
 

dmcowen674

No Lifer
Oct 13, 1999
54,889
47
91
www.alienbabeltech.com
Originally posted by: Ken_g6
Originally posted by: dmcowen674
Best bet like I've said from the beginning is to convince Microsoft to include DC as a drop down Menu option.

Until that happens, DC will be like CB Radios.
Whatever happened to the thing Google was putting in their toolbar? F@H, wasn't it?

IIRC they had to can it because of Feedback due to the fact they did not give you a choice to run it or not, when you downloaded the toolbar, it out the DC program on whether you wanted it or not.

Was a bad approach.
 

Spacehead

Lifer
Jun 2, 2002
13,067
9,858
136
Originally posted by: dmcowen674
Originally posted by: Ken_g6
Originally posted by: dmcowen674
Best bet like I've said from the beginning is to convince Microsoft to include DC as a drop down Menu option.

Until that happens, DC will be like CB Radios.
Whatever happened to the thing Google was putting in their toolbar? F@H, wasn't it?

IIRC they had to can it because of Feedback due to the fact they did not give you a choice to run it or not, when you downloaded the toolbar, it out the DC program on whether you wanted it or not.

Was a bad approach.

Google Compute
The Google Compute feature of the Google Toolbar shouldn't affect your regular computing activities and you can easily disable it at any time for any reason.
Or so it says
 

kranky

Elite Member
Oct 9, 1999
21,019
156
106
Speaking only for myself, one factor is the availability of a DC project that
- runs as a service automatically
- does not require a net connection
- ability to receive work from and send work to a local machine that accumulates the output of all the others
- has a unit of measure which is meaningful. In other words, if a machine works on something for 24 hours, I like to see output that is measured in at least two, preferably three, digits. To work on a project for a whole day and get credit for 3 work units (for example) isn't really satisfying to me. I know that's a personal quirk, but hey. :)

RC5 did all that.

And narzy, first sentence in the third paragraph should be "intents and purposes" not "intensive purposes".
 

BGod

Golden Member
Oct 9, 1999
1,375
39
91
I'd have to agree with kranky on this one. The d.net client is one of the best behaved, easiest to configure, least prone to interfear with other programs client out there. Plus setting up a personal proxy to buffer blocks and have one central machine making the internet connections was a breeze.

My only complaint is that my computers have gotten faster, but yet I'm only putting out one twentyth of the work units daily that I used to. It's amazing how long it's taking me to reach 1,000,000 blocks.

RC5-64 3.34 years = 11,822,304
RC5-72 2.42 years = 912,696