we've noticed over the past several days a dramatic increase in the number of new users per day, which is wonderful! where have new users been finding out about the project? (we've been thinking about how to advertise more widely, and this might provide some ideas).

I've been testing out a number of ways of spreading out the search through possible conformations over the last several days (hence all the jobs with the different names!), and it has been amazing how much data has come in over this time alone. I spent the day going over the ~300,000 predicted structures so far--the "length 10" runs are the best so far. Needless to say, it would have taken months to find this out in the pre BOINC dark ages on our in house computers!

Tonight we are starting runs for a new protein so you will see some new job names. We will be making predictions using the different methods for about ten proteins over the next couple of weeks so we can determine which methods work best.

I see from the other posts that there are still a lot of problems--we still have a lot of work to do! Please let us know what type of computers you are running Rosetta on as this can help us to diagnose the problems.

---

we've noticed over the past several days a dramatic increase in the number of new users per day, which is wonderful! where have new users been finding out about the project? (we've been thinking about how to advertise more widely, and this might provide some ideas).

I've been testing out a number of ways of spreading out the search through possible conformations over the last several days (hence all the jobs with the different names!), and it has been amazing how much data has come in over this time alone. I spent the day going over the ~300,000 predicted structures so far--the "length 10" runs are the best so far. Needless to say, it would have taken months to find this out in the pre BOINC dark ages on our in house computers!

Tonight we are starting runs for a new protein so you will see some new job names. We will be making predictions using the different methods for about ten proteins over the next couple of weeks so we can determine which methods work best.

I see from the other posts that there are still a lot of problems--we still have a lot of work to do! Please let us know what type of computers you are running Rosetta on as this can help us to diagnose the problems.

This was me. Azamaza is my son Benjamin's team, and he has a web site with the same name. I didn't see he was logged in!

---

Obviously we have had a huge number of people just diving in and reading the Wiki ... :)

I suspect, and this is only a suspicion, that part of this is the "honeymoon" effect ... LOTS of interaction between the project team and the participant base... so, those of us that "patrol" the message boards talk about the experiences, good/bad with projects and make recommendations to people that ask questions ...

Now, cynic that *I* am, I am waiting to see how long this will last ... I saw the same kind of early interaction on other projects ... now ... well ...

My other suspicion, is that more of the old "SETI Classic" base is coming over to BOINC and are actually looking at more than just continuing the same old rut ... which is what I also hoped to see ...

Also, you have the advertisements of the signatures ... we "brand" our posts and that also arouses questions ... I guess I am trying to say that it is probably a lot of things ... oh, and I bought another computer ... so that accounts for part of it too ... :)

---

Thanks for the timely feedback. It's great to see the intrest generated by good communications.....Cheers, Rog.

---

(edited this post quite a bit)

Boincstat does show a bit of a surge in the number of 'new users' starting from the 21/22 of this month - it's subtle but noticeable: http://www.boincstats.com/stats/project_graph.php?pr=rosetta

On the other hand the # of 'users per day' shot through the roof when the smaller work units where introduced (as so many more computers were able to return WUs)

As for the 'honey moon' effect - let's hope Rosetta stays immune from it. A strong community will be key.

---

Team CFVault.com
http://www.cfvault.com

As for the 'honey moon' effect - let's hope Rosetta stays immune from it. A strong community will be key.

Well, from *MY* limited perspective, the participant side is pretty homogenous and consistent. It is almost all the same "group" of people.

Those of us that are project "agnostic" and do several projects on all computers tend to do 4-10 projects. I, for example, do all of the production projects including a couple that are still "alpha" rated just to give me a good "mix".

I have also been one who has insisted that there is already a community, to be rediculed in several forums. It is not a well organized community still (Paul's opinion), but, a community none the less. I have also championed, again unsuccessfully, for increased project-to-project cooperation and communication.

If there is much of that, it is hidden, and that is usually not a good thing (again, Paul's opinion, but some say I read too much history too ... now if I could only get a good book on Bio & protein folding, spindling, & mutilating ...).

---

It is possable that Housing and Food Services finally got students signing up? Part would have come from other projects users visiting group web sites and finding out about it. And some changeing over from Predictor as they are much lower in credit than other projects. And we can't forget people using the Waki finding out about it as well as word of mouth.
Maby just a combination of the above with the smaller WU's brought a lot in. Just hope that they all stay with more protiens being analized and all diferant size WU's.
Cheers
Ray

---

Pizza@Home Rays Place Rays place Forums

Maby just a combination of the above with the smaller WU's brought a lot Ray

Yes, the 'jump' seen in user participation is definitely small VUs related. There are also slightly more users signing up every day, but overall it's following the same 'growth curve' as it had before.

---

Team CFVault.com
http://www.cfvault.com