Advanced search

Message boards : General Topics : Some Status Updates & Improving C@H

Author Message
Profile Marius
Project administrator
Project developer
Project scientist
Avatar
Send message
Joined: 29 Jun 15
Posts: 430
Credit: 4,276
RAC: 0
Message 20254 - Posted: 10 Jul 2015, 18:40:32 UTC

Hi everyone,

My name is Marius. Although I've been totally silent on these boards until now, I've been involved in some way or another with C@H for about three years. At first it was just developing PICO (the code which uses the results you guys calculate), more recently full blown C@H development.

I'm here to give you some updates and answer questions. We definitely hear your complaints and admit this project hasn't been maintained well in the past few years. For this we are sorry, and its clear we need to improve to keep you as volunteers. As we see it, the main requests have been,

* Quicker response on these boards
* A clearer explanation of what exactly you guys are calculating and what we do with it
* Proof of the impact of these results (i.e. in papers, etc...)
* Software updates to the C@H apps and server
* No more empty promises

Please tell us if we missed something important. We're going to try and address all of these.

Now for updates. A large part of what I've been doing recently centers around developing boinc2docker, a new app framework which is going to make it infitiely easier for us to deploy future apps, and hence to do more and new science. I will make a detailed post about this in the near future. If you're curious for now you can stalk my github and interactions with the BOINC devs on the boinc_dev and boinc_projects mailing lists (in fact anyone is more than welcome to contribute, we're keeping this all open source).

For now that's it, just wanted to let you know we're actively working on things as we speak, and more will follow. Feel free to leave questions or comments. Thanks!

Marius

Profile cykodennis
Send message
Joined: 31 May 10
Posts: 234
Credit: 4,896,378
RAC: 0
Message 20255 - Posted: 11 Jul 2015, 3:20:37 UTC - in response to Message 20254.
Last modified: 11 Jul 2015, 3:20:56 UTC

Hello Marius,

glad to read you here. I expected your arrival on these message boards later in this year, so your appearance is a great and happy surprise to me.

To talk short about the lack-of-communication complaints - as entitled as they may have been, most of the longtime C@H users should be aware of the fact, that these were mostly the results of the specific circumstances of the project - crew being small, splitted all over the world in different time zones, a very busy main scientist who is involved in many groups and institutions...
And nobody can deny that there were several (unsucessful) attempts made, to fix this problem.
I'm pretty sure that nobody will really think about it any longer, if now work goes on, maybe with new directions and an at least basic communication between crew and volunteers.

I think your list includes the most necessary points.

Imho the most important thing is a better project description, understandable for common people.
Most questions for progress bars, results, publications etc. arose because of the fact, that the people don't really know what C@H does. It is not really explained in Bens "Letter to Cosmology@Home users".
He added some further explanations in the following years, but they are scattered all over the message boards. A new user will face some difficulties when attempting to collect these informations.

Afaik PICO was used in analyzing Planck Data. Maybe this would be a good example for point "proof of impact".

I was really excited when reading the description of boinc2docker on your github (in fact, I'm still excited)! I watched those whole V-BOINC thing since i randomly found a paper of Gary McGilvary in ~2013(?), and i was disappointed when the development went into the virtualbox deadlock.
Guess, i'll have to read a little bit about docker, to be able to ask you smarter questions =)

Again, I'm really happy about your arrival here.
Thank you!

Daniel
Send message
Joined: 28 Aug 07
Posts: 3
Credit: 347,134
RAC: 100
Message 20256 - Posted: 11 Jul 2015, 3:21:12 UTC - in response to Message 20254.

Thanks for your post Marius.
It's appreciated.
Since Ben's last post on the 3rd July, I have started crunching for this project again.
It's good to hear that things are happening.
I look forward to more updates in the future.

Profile Marius
Project administrator
Project developer
Project scientist
Avatar
Send message
Joined: 29 Jun 15
Posts: 430
Credit: 4,276
RAC: 0
Message 20257 - Posted: 11 Jul 2015, 5:38:31 UTC
Last modified: 11 Jul 2015, 5:39:17 UTC

Thanks guys.

Re: PICO, yes! Both Ben and I are members of Planck and PICO is a software which was used extensively in parts of that analysis, and which has directly benefited from C@H.

Re: boinc2docker, I am pretty excited about it too! I saw the V-BOINC paper as well, the idea is good but I'm not sure how far it was ever executed like you say. BOINC itself took it further and has had Virtualbox apps for a while, and our code uses that but goes one step further with Docker. Docker is itself a virtual machine too, but the ease of *building* and distributing Docker machines is where the real advantage is, and why I'm excited.

Message boards : General Topics : Some Status Updates & Improving C@H