Complete "BOINC Unification" in one website

Message boards : Promotion : Complete "BOINC Unification" in one website
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · Next

AuthorMessage
Profile KSMarksPsych
Volunteer moderator
Help desk expert
Avatar

Send message
Joined: 30 Oct 05
Posts: 1239
United States
Message 17050 - Posted: 1 May 2008, 22:12:15 UTC

I agree with Nicolas. The main BOINC site should be about BOINC, not about project science or troubleshooting project issues.

Projects are independent. They're free to modify the backend in any way they want. If they want a "cool and modern" site, they're free to do that. Personally I find the bog standard BOINC pages sufficient for what they need to do. They need to present the information and make navigation easier. Are there things that need improved? Yes. Splitting Forums and QA is dumb. A WYSIWYG editor would be handy for the forums. I've heard adding project news is pretty painful for admins. But the last thing we need is a whole flash based site. Flash is good for a lot of things. But most people equate Flash with cool and modern (a la myspace). Hell, practically every major site in Korea is Flash based. Trying getting Google to translate it. Try to use them in a text based browser. Try to use them with screen reading software. Cool and modern != usable. I find the Orbit front page (no Flash) annoying. I have to click through to join to get to the stand BOINC page to get to my account. Conversations are going on in two places, replies to the "blog" and the forums/QA. How does that make the page better? There are projects with modified front pages that look good yet keep the usability that is needed. Pirates comes to mind as does Docking.

The only thing you're going to get if you put all the project stuff on one site is a lot of very confused people. How many people do we have posting off topic stuff in NC or problems in the cafe? If people can't figure out something that simple, then they'll never figure out which project area to post in.

I see no good coming from this. David brings up the legal issues. I don't blame him one bit for being gun shy. Folks in the US are litigious as hell. Look at the guy who won millions from McDonald's.
Kathryn :o)
ID: 17050 · Report as offensive
Profile yomshleeshee
Avatar

Send message
Joined: 2 Jul 07
Posts: 10
Message 17052 - Posted: 2 May 2008, 0:42:59 UTC - in response to Message 17027.  
Last modified: 2 May 2008, 0:46:24 UTC

Yes, we need a single web site that has complete info on all BOINC projects. That web site should be an account manager...
BOINC is a technology, not a brand. ...


Agreed. That's supported by my earlier post. I have my own brand that I want run on my own. The BOINC system is awesome; so I use it. But I don't see the need to migrate my stuff off of my own server that I can ssh into while sitting at my kitchen table (which I'm doing now ;) ) or at my workstation at the university. And I reiterate that I don't see what's missing from BAM! or GR plus a good team (eg Team Picard...). I see that as 2 links I click every day. Between all of us in the team, news gets posted. Thanks to Willy, I can manage all of my hosts and project participation in one spot. I'm just not seeing that anythings missing. Frankly, not to bring down John's suggestion (glad he's so active), I can't see myself sifting through all the threads and fora that would be piled on in one heap if all of the project were in one message board. My real opinion is that the structure should stay the same. There are some things that need to be fixed. That being said this is open source. So if it doesn't get fixed, or as is often the case with me, things get changed, but not checked back in, that's our fault as project managers. But not because there's no unification. I mean I know I've put some patches on the "brats_ops". And I think Kathryn's got some good suggestions (as always). And to be honest, I'm not sure this is what is meant or not, but if project hosting was moved to one place (like WCG?) and I could not host myself, I'd probably bail on the whole thing. Not sure if the suggestion meant for that, but just throwing that out there. That's my 2 cents (again). For what that's worth.

PS. not trying to bring John and his ideas down. Glad his thinking about it and throwing himself out there. Just don't like the idea. Sorry.

Team Picard
ID: 17052 · Report as offensive
Profile yomshleeshee
Avatar

Send message
Joined: 2 Jul 07
Posts: 10
Message 17053 - Posted: 2 May 2008, 0:48:47 UTC

Oh and another thing. BOINC page has done a good job of providing a unified page for future or developing project admins to create and manage their projects. In fact these have seemed to have grown since I came around... (eric meyers is particularly good). So there is an example of there being one spot for admins to go for help. That already exists too.

Team Picard
ID: 17053 · Report as offensive
Profile Mike O

Send message
Joined: 1 May 08
Posts: 24
United States
Message 17054 - Posted: 2 May 2008, 1:22:16 UTC
Last modified: 2 May 2008, 1:31:14 UTC

WOW.. vary different opinions here.
ok..
Flash. like it or not is used by almost EVERY major site in the USA and the world. NOT just Korea. If you cant deal with it, your computer needs an UPGRADE!
If you think for one second I want BOINC to look and feel like you just stepped in to a DISC in the mid 70's.. you are more far off than Pluto!
Also, the way my single flash button works, the text and link is loaded in to it via PHP.. that button dont give a hoot what language it is! It would not be real hard to write a script that could change ALL the text to any language. I how ever speak english and my site is for testing more than any thing. Im not managing anything but my code there.. lol.
I stated in the seti forums that the overall design should be decided on.
A poll, an email adress to comment to.. I DONT CARE but I think the general type of coding used beit flash, css or regular old html dont matter.. many things can be done with html and php.. like all the forums done in php that are spread all over the world about the projects..ALL based on html/php... and thats the problem here that I see most of you ARE NOT GETTING!!
Is it true or false to state that BOINC is 'THE FATHER/MOTHER' of all the projects?
Lets think about this for a second.
With out it.. NO PROJECTS would be accessible! DO you want BOINC to simply be an application that lets you connect to ANY project thus opening the door for some crazed hacker to setup a bogus project and destroy your system or WORSE!
BOINC needs and must have the ability to manage ALL projects to guard this from happening. BOINC should be the SINGLE account manager for ALL you private information including a client/user ID that MUST be matched by ANY project before it is allowed to be downloaded. ALL projects MOST be varified by BOINC to make sure they are REAL PROJECTS!

I have said this before.. If someone cant figure out how to simply join a project, they need to go back to surfing porn! They have NO I.Q. what so ever and could not pour sand out of a boot with the instructions on the heel! People that skilless are the types that will download a project to 'check it out' and uninstall it leaving uncompleted WUs waiting to time out for weeks.. A TOTAL WASTE OF TIME for them and the serious crunchers!

This is one of the main reasons I am totaly for a boinc unification! People need to be made VARY AWARE of what the intentions of BOINC AND THE PROJECTS are and that this is not something to sign up for for 'kicks' just to uninstall it in a week or two.
Right now, the state of boinc is... a bunch of people running what ever they want to play 'who can get to the top first' It's a freaking contest to see who has the 'biggest computer' Kinda childish IMHO. Do I care about my stats? Well.. I really want to know if there is life out there and so yes.. the more I can get done, the better. I put my 'farm' together to do 3D rendering but decided this is more useful than playing around with 3D graphics.
I really do think most dont give a crap about the science!! Why? because, BOINC DONT CARE WHO SIGNS UP!! Reminds me of the company I work for.. Hire anyone... as long as they show up. The company doesn't seem to care how much machining is screwed up however. Sure there are some that really DO care... as it is with BOINC.. some care where this goes.. want to see improvements... expansion... as I do with the company I work for and BOINC.
Stats.. theres that d*** contest again.. I dont care about who has what, when or where.. I WANT TO SEE SOME RESULTS OF THE DATA THAT IS BEING PROCESSED!!! To be honest, it is very anoying to have to scroll through a single PAGE of someones stats to read the next post! I think stats that large should be out right banned.

We have forum moderators for every single forum of every single project but don't seem to have any moderators of the projects themselves. Why and how could something that simple have been over looked????
I have so many ideas running threw my head about the usefulness of a unified boinc I cant seem to think straight.

One thing that could be done...
For the 'lookie loos' out there that just wanna signup to 'check it out'...
A new core version of BOINC that will NOT let it be uninstalled until the current work unit is completed. The core would remain installed but missing from the uninstall list but when triggered by the uninstall attemp, would no longer request any work units.. signal the project's server what units it will NOT be completing in the cache.. hide its self.. finish the current one it started and then remove the core version from the users system completely. I would also make it clear to the user during the unistall attemp exactly what I just stated above. This is all for windows users.. I know nothing about unix users or how any of that works.. I do know that unix users tend to be much more 'into' computers than the average user.

Myspace... DISCO.. Weeeeeeeeee!! Lets BOOGIE!!
Now really... The sites need a make over.. Seti and most the sites I have seen in the projects look like they were a junior highschool students homework assignment. I understand that the scienctist are not here to 'make web sites'. They are here to oversee the science, right? They dont have the time or man power for it.
Every project can do what ever they want... thats cool, until you want to navigate and find that there are no buttons on one site to go 'there' that are at the other site. You go from one site thats bright, upbeat to the next that's dull, black and white and just plain boring. Help info is at the bottom at one site and ohmmm... 'can't find it' at another's.
One of the key things is the univeral use and looks of the unified site. A standardized Menu bar would be established and ALL pages would adhear to that Menu Standard. A common color scheme or 'Motif' if you will for all pages.
And most of all.. SCIENCE and lots of it! I think a policy of baring projects that can not or will not return usable science results should be banned from the BOINC community! Each project will have its own page(s) to display their science but MUST follow the standardized menu and color motif although, they can place the projects banner on the pages in any color scheme they chose but must be of the same physical diminsions, occupying the same location common to all project pages.
A mod to the BOINC core could KILL projects on the client's side along with a message to that client's user explaing exactly why it was killed. "Project 'whatever' has been deemed useless and is now being removed from BOINC for not following the "BOINC Project Guide Lines" To read more, click here" or something similar to that.

A customizable "user's home page" where they can list the science from all there projects they are crunching, none or the ones they chose. Im invisioning something like mnay sites have where you can place and remove things you like.

A forum JUST for coders and designers.. which BTW.. you have to pay 1.00 dollar/month to be involved with. Its a buck a month (big deal huh?) and simply by having to get an account set up, will keep most of the idiots out that are looking to disrupt things. The procedes could maybe help fund something useful like a new server just for the Unified BOINC site?

John.. I like every thing you have said.. one thing to be careful about.. to many in charge.. A absolute must is an organizational flow chart of who is at the top, working down to the bottom showing who has authority over others is very important. for example.. Thoses that give a buck to be coding need not be worring about who chose the Menu to be neon yellow. They just need to focus on making the menu or working with the coders assigned that task. (neon yellow... think im gonna be sick now.. lol)

Anyway.. this turned in to a rather large post.. I so hope that the ones that believe chaos is better then orginzation, start to understand what a 'Unified Boinc' is about.. BOINC wont design Projects but they will make F'ing sure they 'play by the rules' and RETURN SCIENCE!!! This is after all... why it was created, isn't it?

Thanks for reading...
Mike
(note: Add to things to do for UB "A SPELL CHECK SCRIPT" lol)
ID: 17054 · Report as offensive
Eric Myers
Avatar

Send message
Joined: 12 Feb 06
Posts: 232
United States
Message 17055 - Posted: 2 May 2008, 1:52:06 UTC - in response to Message 17046.  

Legal trouble or not, I agree that the official BOINC website shouldn't have project-specific stuff.

Well, I have to disagree. See the thread nextdoor. The point being that the main BOINC site could be a way for projects to share general news, information about what's going on (the outages thread is very useful) and generally stay connected. It's not a requirement, but it could be useful.

...BOINC website shouldn't have detailed explanations problems.

I'll agree that the project-specific details should be left to the project web sites, and projects should not expect some kind of central service from BOINC. They actually have to set up their own site for their participants.

But these forums could provide a method of inter-project communication, in addition to or instead of an e-mail list.

-- Eric Myers

"Education is not the filling of a pail, but the lighting of a fire." -- William Butler Yeats
ID: 17055 · Report as offensive
Profile yomshleeshee
Avatar

Send message
Joined: 2 Jul 07
Posts: 10
Message 17056 - Posted: 2 May 2008, 1:53:56 UTC - in response to Message 17054.  
Last modified: 2 May 2008, 1:55:21 UTC

Couple of comments to keep the debate going:


...that this is not something to sign up for for 'kicks' just to uninstall it in a week or two.

I don't know about other project admins out there. But I haven't seen this to be a problem. Yes there are some that drop out, but 1)most people stick around for at least 1 WU and 2)if you are loosing WU crunching because of people signing up, downloading and then detaching, you need to bump up your Result quota.

I really do think most dont give a crap about the science!!

I only agree slightly. I think most people do care about the science. Yes there's more credit fighting than I care for, but I'd give more credit to the participants.

A new core version of BOINC that will NOT let it be uninstalled until the current work unit is completed.

Seriously? You really want an application to refuse to uninstall itself. That seems like a whole 'nother can of worms.

I understand that the scienctist are not here to 'make web sites'. They are here to oversee the science, right? They dont have the time or man power for it.

+1000

A mod to the BOINC core could KILL projects on the client's side along with a message to that client's user explaing exactly why it was killed.


Don't like this at all. Hacking aside. Who should decide what project should or shouldn't get killed. Bad idea. That's what teams are for.

As for the website comments, are they really that difficult? May be I am a nerd. I just don't see the need for flash on a project's website. Am I trying to attract people based on features and flashy text? I thought you made fun of that earlier. Anyways, don't want to turn this into a web design debate. I guess maybe the main point is uniformity in the web pages' appearance. I understand it's easier to find this if you know where you're looking, but as for me, I like that each project has it's own look.

Team Picard
ID: 17056 · Report as offensive
Profile Mike O

Send message
Joined: 1 May 08
Posts: 24
United States
Message 17057 - Posted: 2 May 2008, 1:56:36 UTC

Some more thoughts here.

Before a project is granted a slot in the Unified Boinc site. There proposed project and discription of said will be place in the forums for all to read and study. A poll will be taken of ya's and na's and the majority will rule.

Surely, you know all this is just speculation of what could be done.. not what WILL be done.. its a democracy and it should be polled before the end users before implmenting any changes. Once again.. majority rules.

Im outter here for the night.. Morning arrives way faster than the work day goes by :)
wishing you all peace of mind...
Mike
ID: 17057 · Report as offensive
Profile yomshleeshee
Avatar

Send message
Joined: 2 Jul 07
Posts: 10
Message 17058 - Posted: 2 May 2008, 2:11:35 UTC - in response to Message 17057.  

Before a project is granted a slot in the Unified Boinc site. There proposed project and discription of said will be place in the forums for all to read and study. A poll will be taken of ya's and na's and the majority will rule.


I mean I know that it would probably be rare for something to get denied. But who's going to decide what project does or doesn't get run. Big lab versus small lab is why I'm here and not on my university's Cluster in the first place. I would really hate to see BOINC being micromanaged. Also, I know the awesome guys at Berkeley have enough to do with out having to be on a BOINC project committee. General public instead? Hmmm...

Surely, you know all this is just speculation of what could be done.. not what WILL be done.. its a democracy and it should be polled before the end users before implmenting any changes. Once again.. majority rules.


Of course, of course. That's why we're debating it.


Team Picard
ID: 17058 · Report as offensive
Profile Mike O

Send message
Joined: 1 May 08
Posts: 24
United States
Message 17059 - Posted: 2 May 2008, 3:39:46 UTC - in response to Message 17056.  

Couple of comments to keep the debate going:


...that this is not something to sign up for for 'kicks' just to uninstall it in a week or two.

I don't know about other project admins out there. But I haven't seen this to be a problem. Yes there are some that drop out, but 1)most people stick around for at least 1 WU and 2)if you are loosing WU crunching because of people signing up, downloading and then detaching, you need to bump up your Result quota.

I really do think most dont give a crap about the science!!

I only agree slightly. I think most people do care about the science. Yes there's more credit fighting than I care for, but I'd give more credit to the participants.

A new core version of BOINC that will NOT let it be uninstalled until the current work unit is completed.

Seriously? You really want an application to refuse to uninstall itself. That seems like a whole 'nother can of worms.

I understand that the scienctist are not here to 'make web sites'. They are here to oversee the science, right? They dont have the time or man power for it.

+1000

A mod to the BOINC core could KILL projects on the client's side along with a message to that client's user explaing exactly why it was killed.


Don't like this at all. Hacking aside. Who should decide what project should or shouldn't get killed. Bad idea. That's what teams are for.

As for the website comments, are they really that difficult? May be I am a nerd. I just don't see the need for flash on a project's website. Am I trying to attract people based on features and flashy text? I thought you made fun of that earlier. Anyways, don't want to turn this into a web design debate. I guess maybe the main point is uniformity in the web pages' appearance. I understand it's easier to find this if you know where you're looking, but as for me, I like that each project has it's own look.


Like I said in my last post.. NONE of this is real or even close to happening, these are ideas from my gray matter and no one elses.

About Flash... Flash has many MANY uses.. you do NOT have to have graphics on a page to take advantage of its usefulness but it could be used as a powerful tool to represent data.

John threw it out there.. an 'idea' for a more integrated world for the project participants and the projects themselves. This will never happen because it means rebuilding the entire BOINC system from the floor up. I see NO government over what is considered real science and a waste of clock cycles. I heard there was a project someone started looking for the largest prime numbers? Ok.. why? Why not do one to find the longest solution to PI. They are both as valuable.
In the fear of having this post deleted I will say it anyway...
I thought the whole point of BOINC was to distrubute science so the public could get involved. How many people in the world to you imagine needs to know the value of PI to, say, 100 or 1000 decimal places? In my opinion, that isnt science... its just number crunching and has no business trying to fool people in to thinking other wise or diverting people from doing real and useful science. I also think it was a MAJOR mistake making boinc so open that even hackers are f'ing with people let alone the useless, so called science.

I wonder if any project admins have parsed their database to see how many have signed up only to stop crunching after one, two or three WUs. Id be interesting to know what percentage they are verses the average users life span.

The main idea I was trying to get at with the 'web page stuff' is that there is little or NO science displayed that the average Joe can understand or get excited about so they quit.

Im tired.. its late here.. after 11:00 pm now and I get up at 5:30 am but before I go to bed, I will check your project and your site.. I have stuck with seti because it has always been a burning issue in the back of my mind ever since I watched Carl Sagan's "Cosmos".

You my friend are a thinker and I can apreiciate that. I knew when I posted about boinc being able to kill a project, it would raise a few eyebrows..
I just feel like this whole project could be done better.. As in science that could change the world for the better.
If only the funding was there, I know none of us would be posting about any of this. There really is true genius in the staff that put this together!

ID: 17059 · Report as offensive
John37309
Avatar

Send message
Joined: 28 Jul 07
Posts: 91
Ireland
Message 17060 - Posted: 2 May 2008, 3:40:30 UTC

Taking into consideration what David has said already, here is one possible solution that might help with better unification.

The idea is called "BOINC-LINK" and it is a set of images that are displayed in the <header> of the participating websites. Each of the images is a link to the other website. So with very simple ease, you can hop from one website to the other.

Many other websites do something very similar to this. Once David has agreed on the websites that he is willing to support as "BOINC Help and Support" websites, all of the selected websites would agree to display an identical set of "BOINC-LINK" images in the <header> of EVERY page on each of the websites.

This is what the set of images might look like with some of the current websites like BOINC Berkeley, BAM, GridRepublic, Boincteams.com and BOINC-WIKI;



But the set of images could include more, less or other websites like this one;



This "BOINC-LINK" system would only work if David is willing to support the idea. David would also have to attempt to get this into the <header> of his website, wiki, trac and any other pages on the site. Its not difficult. But equally, the other participating websites would have to do the same.

Space.com is the only example i can think of that has something kinda similar to what this looks like and how it might work.

Thoughts anyone?

John.
| Irelandboinc.com | PaddysInSpace.com
ID: 17060 · Report as offensive
Profile Mike O

Send message
Joined: 1 May 08
Posts: 24
United States
Message 17061 - Posted: 2 May 2008, 3:44:19 UTC - in response to Message 17058.  

General public instead? Hmmm...


Yes!! let the people that will be crunching the data decide..
POWER TO THE PEOPLE!!
and think of how that would impact everyone by having the ability to actually VOTE on a projects exceptence!

ID: 17061 · Report as offensive
Profile KSMarksPsych
Volunteer moderator
Help desk expert
Avatar

Send message
Joined: 30 Oct 05
Posts: 1239
United States
Message 17062 - Posted: 2 May 2008, 3:44:40 UTC - in response to Message 17054.  
Last modified: 2 May 2008, 10:20:13 UTC

WOW.. vary different opinions here.
ok..
Flash. like it or not is used by almost EVERY major site in the USA and the world. NOT just Korea. If you cant deal with it, your computer needs an UPGRADE!
If you think for one second I want BOINC to look and feel like you just stepped in to a DISC in the mid 70's.. you are more far off than Pluto!


I never said my computers can't handle Flash. I said it's a nightmare for accessiblity and people who choose to use a text based browser (not me, but there are people).

Flash is way over used. Flash != good web design. Common != GOOD.

Are the BOINC web pages pretty? Not really. But are they functional? Yes, for the most part.

Lately, there's been way too much emphasis on the "social networking" stuff. I refused a friend request for someone and got an abusive email back from that person. That's not why I signed up for BOINC. I signed up because I'm interested in the science, I'm interested in what DC can do, and I want to learn more about technology.



I stated in the seti forums that the overall design should be decided on.


But how many people visit the forums on a regular basis? How many have the skills to participate in rewriting the webcode? And how many give a hoot? I'm not seeing a problem with the web pages.


forums done in php that are spread all over the world about the projects..ALL based on html/php... and thats the problem here that I see most of you ARE NOT GETTING!!


You're right. I'm not seeing that there is a problem. What is wrong with the html/css/php we have? Are there issues. Yes. I acknowledge that before. But let's fix the bugs and enhance them rather than starting over.

Did you read the debate on Seti when the host page was changed? People didn't want it to change. You'll have just as many people screaming when the whole web design changes. It's come to the point that I have images and sigs turned off on all the forums except those where I'm a mod. Flashing smiles and 12 inch signatures aren't my cup of tea.


With out it.. NO PROJECTS would be accessible!


I agree to some extent. There are other DC frameworks out there. If not BOINC, than someone else will do it.

DO you want BOINC to simply be an application that lets you connect to ANY project thus opening the door for some crazed hacker to setup a bogus project and destroy your system or WORSE!


It's already this. The potential is there. Just because it hasn't happened, doesn't mean that it can't or won't happen.



BOINC needs and must have the ability to manage ALL projects to guard this from happening. BOINC should be the SINGLE account manager for ALL you private information including a client/user ID that MUST be matched by ANY project before it is allowed to be downloaded. ALL projects MOST be varified by BOINC to make sure they are REAL PROJECTS!


BOINC operates on a limited budget with three full time programmers. Who is going to verify the projects? What about projects that don't publish their source? This is a value judgment and who's the say David's/Rom's/Charlie's values are correct (not that I don't trust David/Rom/Charlie).

I have said this before.. If someone cant figure out how to simply join a project, they need to go back to surfing porn! They have NO I.Q. what so ever and could not pour sand out of a boot with the instructions on the heel! People that skilless are the types that will download a project to 'check it out' and uninstall it leaving uncompleted WUs waiting to time out for weeks.. A TOTAL WASTE OF TIME for them and the serious crunchers!


This is totally uncalled for. When I started out, I found the entire thing confusing. I learned though. And the people who are really interested will ask for help or learn on their own. Computer unsavy != retarded (which is exactly what you're saying).

This is one of the main reasons I am totaly for a boinc unification! People need to be made VARY AWARE of what the intentions of BOINC AND THE PROJECTS are and that this is not something to sign up for for 'kicks' just to uninstall it in a week or two.


This is something that should be made clearer on the Berkeley site. However, you don't need to unify all the projects to do this. We need to improve the documentation. There are people working on the wikis now. Perhaps this is an area you can work in.


Right now, the state of boinc is... a bunch of people running what ever they want to play 'who can get to the top first' It's a freaking contest to see who has the 'biggest computer' Kinda childish IMHO.


Different strokes for different folks. I like my stats, but I don't live and die by whether my world ranking goes up or down. Calling the credit hounds childish is pretty insulting to them.

I really do think most dont give a crap about the science!!


So? What difference does their reason for running BOINC make?

Why? because, BOINC DONT CARE WHO SIGNS UP!!


And why should BOINC care?



One thing that could be done...
For the 'lookie loos' out there that just wanna signup to 'check it out'...
A new core version of BOINC that will NOT let it be uninstalled until the current work unit is completed. The core would remain installed but missing from the uninstall list but when triggered by the uninstall attemp, would no longer request any work units.. signal the project's server what units it will NOT be completing in the cache.. hide its self.. finish the current one it started and then remove the core version from the users system completely. I would also make it clear to the user during the unistall attemp exactly what I just stated above.


BAD IDEA!!!!!!!!!!!!! That's basically malware.


Every project can do what ever they want... thats cool, until you want to navigate and find that there are no buttons on one site to go 'there' that are at the other site. You go from one site thats bright, upbeat to the next that's dull, black and white and just plain boring. Help info is at the bottom at one site and ohmmm... 'can't find it' at another's.


So let's modify the standard BOINC package. Again, I'm not saying the front page on the standard server is pretty. But it does have all the information a person needs.


One of the key things is the univeral use and looks of the unified site. A standardized Menu bar would be established and ALL pages would adhear to that Menu Standard. A common color scheme or 'Motif' if you will for all pages.
And most of all.. SCIENCE and lots of it! I think a policy of baring projects that can not or will not return usable science results should be banned from the BOINC community! Each project will have its own page(s) to display their science but MUST follow the standardized menu and color motif although, they can place the projects banner on the pages in any color scheme they chose but must be of the same physical diminsions, occupying the same location common to all project pages.


A unified site and a unified style are two completely different things. I don't have such a huge problem with a unified style. But you shouldn't make it mandatory. Seti doesn't even use the supplied front page.

A mod to the BOINC core could KILL projects on the client's side along with a message to that client's user explaing exactly why it was killed. "Project 'whatever' has been deemed useless and is now being removed from BOINC for not following the "BOINC Project Guide Lines" To read more, click here" or something similar to that.


Again, who is to determine what is "useless". Searching for mega primes? Searching for ET? Searching for gravity waves? Predicting protein structure? That's a judgment call. Personally I find Seti the biggest waste of CPU power out there. But others would say the same about PrimeGrid.

A customizable "user's home page" where they can list the science from all there projects they are crunching, none or the ones they chose. Im invisioning something like mnay sites have where you can place and remove things you like.


Meh. Eye candy. I don't care for it. But that doesn't mean there aren't people who do (look at all the MySpacers).

A forum JUST for coders and designers.. which BTW.. you have to pay 1.00 dollar/month to be involved with. Its a buck a month (big deal huh?) and simply by having to get an account set up, will keep most of the idiots out that are looking to disrupt things. The procedes could maybe help fund something useful like a new server just for the Unified BOINC site?


What's wrong with the mailing lists and the forum here? Again, I see you seeing problems when there really aren't any. There is very little spam here and almost no trolling here. The mailing lists are the same.


Anyway.. this turned in to a rather large post.. I so hope that the ones that believe chaos is better then orginzation, start to understand what a 'Unified Boinc' is about.. BOINC wont design Projects but they will make F'ing sure they 'play by the rules' and RETURN SCIENCE!!! This is after all... why it was created, isn't it?


As is my reply. But I feel that a Unified site would bring more chaos. I'm not sure where you're seeing the chaos (unless it's the Seti forums) because the vast majority of the projects have few issues.

I've said it before to the developers. We don't need to throw technological solutions at a social problem. We need to fix the social stuff. A flashy website isn't going to stop people from bailing out of BOINC. I lost interest in Seti Classic in under a year. BOINC has kept my attention for 2.5 years now. I've met good people and I've learn a whole.

(note: Add to things to do for UB "A SPELL CHECK SCRIPT" lol)


LOL. One word. Firefox.
Kathryn :o)
ID: 17062 · Report as offensive
Profile Mike O

Send message
Joined: 1 May 08
Posts: 24
United States
Message 17063 - Posted: 2 May 2008, 3:54:07 UTC - in response to Message 17060.  

Hey John.. very good idea.. very much like what I was talking about. Having a common menu on all pages.. what about mini graphic versions of the logos that are resting on a bar of link buttons. Buttons take NOTHING to write and the code is very tight. Could even do a rollover so maybe some text pops up with a tight little discription or larger versions of the logos?
Just an idea :)
Peace.
ID: 17063 · Report as offensive
John37309
Avatar

Send message
Joined: 28 Jul 07
Posts: 91
Ireland
Message 17069 - Posted: 2 May 2008, 11:56:08 UTC - in response to Message 17060.  
Last modified: 2 May 2008, 11:58:55 UTC

Taking into consideration what David has said already, here is one possible solution that might help with better unification.

The idea is called "BOINC-LINK" and it is a set of images that are displayed in the <header> of the participating websites. Each of the images is a link to the other website. So with very simple ease, you can hop from one website to the other.

Many other websites do something very similar to this. Once David has agreed on the websites that he is willing to support as "BOINC Help and Support" websites, all of the selected websites would agree to display an identical set of "BOINC-LINK" images in the <header> of EVERY page on each of the websites.

This is what the set of images might look like with some of the current websites like BOINC Berkeley, BAM, GridRepublic, Boincteams.com and BOINC-WIKI;



But the set of images could include more, less or other websites like this one;



This "BOINC-LINK" system would only work if David is willing to support the idea. David would also have to attempt to get this into the <header> of his website, wiki, trac and any other pages on the site. Its not difficult. But equally, the other participating websites would have to do the same.

Space.com is the only example i can think of that has something kinda similar to what this looks like and how it might work.

Thoughts anyone?

John.

Einstein@home has 4 of these images on the top of the front page. Unfortunately, they have not done the images very well as they are different sizes and the sites they link to have not properly reciprocated the linking images in reverse to link back. But you can get the idea of how it might work!

An agreed size and proportion for the images would be agreed on and we could make the set of images look very professional.
| Irelandboinc.com | PaddysInSpace.com
ID: 17069 · Report as offensive
FilmFreak

Send message
Joined: 9 Oct 06
Posts: 6
Netherlands
Message 17076 - Posted: 2 May 2008, 15:43:42 UTC

I think this bar is a very good idea, but perhaps the complexity of the BOINC manager and BOINC itself is what puts most people off. Feature after feature is added and with every feature, BOINC gets more complex.

I'd really think we need a pop-up when BOINC is first started, asking the user if it wants to view information about:

- BOINC (http://boinc.berkeley.edu/wiki/index.php/User_Manual)
- The BOINC manager (http://boinc.berkeley.edu/wiki/index.php/The_BOINC_Manager)
- The BOINC credit system (http://boinc.berkeley.edu/wiki/index.php/Computation_credit)
- Available projects (http://boinc.berkeley.edu/projects.php)
- Account managers (http://boinc.berkeley.edu/wiki/index.php/Account_managers)
ID: 17076 · Report as offensive
Eric Myers
Avatar

Send message
Joined: 12 Feb 06
Posts: 232
United States
Message 17078 - Posted: 2 May 2008, 16:39:58 UTC - in response to Message 17060.  

Space.com is the only example i can think of that has something kinda similar to what this looks like and how it might work.

This sounds similar to a web ring.

I can imagine projects might put these links at the bottom of their first page, but not at the top.

-- Eric Myers

"Education is not the filling of a pail, but the lighting of a fire." -- William Butler Yeats
ID: 17078 · Report as offensive
Profile yomshleeshee
Avatar

Send message
Joined: 2 Jul 07
Posts: 10
Message 17079 - Posted: 2 May 2008, 16:44:48 UTC - in response to Message 17078.  
Last modified: 2 May 2008, 16:45:29 UTC

This sounds similar to a web ring.

Funny I was thinking the same thing.

I can imagine projects might put these links at the bottom of their first page, but not at the top.


Agreed. I would. Not sure I would want that to be at the top of my page. Again, that's getting to web design. But to validate the point, I would prefer it at the bottom. If I included it at all, depends how the final product look. But I definitely would prefer a bar to a unified BOINC.

Team Picard
ID: 17079 · Report as offensive
erebus

Send message
Joined: 2 May 08
Posts: 1
United States
Message 17080 - Posted: 2 May 2008, 17:13:48 UTC - in response to Message 17062.  

A mod to the BOINC core could KILL projects on the client's side along with a message to that client's user explaing exactly why it was killed. "Project 'whatever' has been deemed useless and is now being removed from BOINC for not following the "BOINC Project Guide Lines" To read more, click here" or something similar to that.



Again, who is to determine what is "useless". Searching for mega primes? Searching for ET? Searching for gravity waves? Predicting protein structure? That's a judgment call. Personally I find Seti the biggest waste of CPU power out there. But others would say the same about PrimeGrid.


I have been watching the debate a little and figured I would throw in my points. I am personally new to the whole being a scientist, (pardon me while I give my qualifications) I spent 4 years in undergrad learning to be a scientist (physics and earth & space science), followed by 2 years of graduate work for my masters (physics) and now I am 1 year into my PhD program.

begin{rant}

One thing I have learned is that even with all my classes and all the time I have spent looking at my own specific research I really do not understand the higher level science research that other people are working on, the same way that they do not understand mine.

A few other things I would like to point out. First of all, BOINC (for the majority of projects) is used to test a theory, the work units are sent out and crunched to see if something that is not accepted is true or not (I stress in most cases). There are many projects that are highly specific to research and only a handful of people in the world will even understand what is being done. I personally can not use BOINC for my research, however I do have to write papers that are reviewed by professionals in the field before they are published to keep unconfirmed science out of the true press (I say true press meaning not the American popular press). This would be the real place that can judge if the science done was valid or not.

Until the paper is reviewed the science holds no true ground. How could someone who may or may not understand the science really decide if it is valid or not with out the intimate knowledge the person who designed the project has? To be honest, in my own research group there are probably 15 people in the world that have access to my data reduction programs and can understand all the steps that they are doing. Does that mean that someone who may be a geologist should be able to say "your astrophysics programs don't run valid science so you have to stop" when they have no idea what I am planning on publishing? There are so many avenues that I could take with my final publication that it is impossible to say if the data processing is valid or not.

So here is my point, there are governing bodies that are already in place in the science community to review the reduced data and ensure that the results are valid. If you put someone above the scientist, then you are adding bureaucracy that does not need to be there and has the potential to destroy science (I know a lot of Americans want to see science come to an end, I feel persecuted in some circles because I am a scientist).

I would never crunch a work unit again if someone else was judging the projects and deciding what science gets done.

Also note (and this probably should have been my first point) that for the project to be started/programmed in the first place someone had to have funded and approve the project. That means that a grant or feasibility study was done to prove to someone who understands the science that there is valid science that will be gained.

If you are not a scientist, don't try to ruin someone’s research by trying to provide a way of deleting it when you do not understand it.

end{rant}


On another note, sure I love pretty web pages, but they are a waste of time. Yes eye candy is important, but to be honest...I hate flash pages. I hate to wait for a little flash thing to load and then I have to over one pixel to get a menu to pop up that then disappears when I try to hover over another item.

Anyhow, sure upgrade the front end if you want so that it is pretty, but do not sacrifice the science because people do not understand.

ID: 17080 · Report as offensive
stwainer

Send message
Joined: 2 May 08
Posts: 2
United States
Message 17086 - Posted: 2 May 2008, 18:16:56 UTC
Last modified: 2 May 2008, 18:18:02 UTC

Try this: Blog

It's just a rough outline of a very, very simple way to have "all BOINC" stuff be available in one place.

Steve
ID: 17086 · Report as offensive
mo.v
Avatar

Send message
Joined: 13 Aug 06
Posts: 778
United Kingdom
Message 17087 - Posted: 2 May 2008, 18:29:47 UTC

David Anderson has said that the only projects that can be included on the 'Boinc-approved' list are those which he knows have all the security precautions in place and where he knows the main administrator.

There are some Boinc projects out there whose administrator's/s' real names and locations are nowhere to be found, at least by searching from their projects' index pages. In some cases I've been unable to determine why a project has been set up or what its tasks attempt to do apart from distributing credits. This has not, of course, deterred thousands of people from attaching and crunching, but could one recommend such projects in an undifferentiated list?

I'm surprised, however, that more apparently very respectable projects haven't been added to the attach wizard.
ID: 17087 · Report as offensive
Previous · 1 · 2 · 3 · 4 · Next

Message boards : Promotion : Complete "BOINC Unification" in one website

Copyright © 2017 University of California. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.