The technical news is a great idea and the new screen saver looks great. The stats idea is great also, good job!

---

Yes, a very very BIG THANK YOU indeed to the team - the screensaver is great! I especially love the explanation page. The site's additions are also very welcome. Brilliant!

---

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

Wow ! The new 'Quick guide to Rosetta and its graphics' seems to be a great introductory text for first time Rosetta users/first time visitors of the site (particularly the first part) and I think it deserves a prominently placed link on the main page. I guess I wouldn't necessarily look for such a great 'Rosetta in a nutshell' type text under its current link called 'Rosetta Graphics'. It is a pity the full page title ('Quick guide ...') may be too long for the main page link (also note the two typos: thier -> their, refered -> referred). I also like the improved 'Server Status' section with the neat little links and the additional logos at the top make the page look really impressive (Rosetta may be fun but this is also important and serious stuff which we are participating in).

I fixed the typos.

I personally think the site overall is just great :)

really well put together, nice layout, warm and welcoming, quite a bit of info (more than i can understand anyway)

it just has a really good feel to it, unlike even seti's revamped site, but then they've got other things to worry about at the moment ;)

i like the updates and upgrades.. a nicer lookin site attracts more people

and the screansaver is enjoyable to run also

---

i like the updates and upgrades.. a nicer lookin site attracts more people

very true, not only is there a lot of information, not only about the science, but how the project (from a tech and boinc point of view) is progressing, but it's presented very well too, this is hard to say, as i've been a dedicated seti cruncher for many years, but i do actually think this site far exceeds the standards of any other, and is much easier and more enjoyable to use and navigate

and the screansaver is enjoyable to run also

i totally agree, i've been watching a few runs lately, and it's facinating, just a bit worrying that most of mine look nothing like they're ment to, i think some, high-level imporvements need to be made about the seach decision making process, because i did have one that was quite close, then it started changing the structure completely, and ended up nothing like it way, surely it should reaslise "ok, this is very different, and not as good as before, so i think i'll go back to where i was before and keep trying" but if i does get to a point where a certain folding method is a best as it can be, only then try something radical to try to get lower, i've seen some folds that look totally different, but still have a low energy level, but as with everything, i'm guessing this is not an easy problem, i'm no programmer, so i'm not going to make any assumptions, i'm merely stating observations

... just a bit worrying that most of mine look nothing like they're ment to, i think some, high-level imporvements need to be made about the seach decision making process, because i did have one that was quite close, then it started changing the structure completely ...

You have, in this short quote, put your finger EXACTLY on why we're doing this distributed processing style. Someone once described it like this. "You've got a gravel pit full of gravel, with two or three uncut diamonds in it. You need to look very closely at each chip of rock in there to see if it's a diamond. How do you do this quickly?"

If you have a million rock chips in the gravel pit, give ten each to one hundred thousand people, and you'll find the diamonds in no time.

To put that another way, we're going to have a lot of time when the protein is way off in left field. That's the nature of the beast we're dealing with.

---

love that analogy dgnuff.

remember in projects like this a left field result is almost as valuable as a close to the mark one since they can narrow down the serch.

---

remember in projects like this a left field result is almost as valuable as a close to the mark one since they can narrow down the serch.

ah, so my results are actually useful?
are they trying to improve rosetta along the way, or are they just trying to get science results from it?

remember in projects like this a left field result is almost as valuable as a close to the mark one since they can narrow down the serch.

ah, so my results are actually useful?
are they trying to improve rosetta along the way, or are they just trying to get science results from it?

Useful, only in that they remove some gravel from my hypothetical gravel pit above, and help the team as a whole find the diamonds. But that is the way with these. The observation needs to be made that if they could refine the algorithm to the point where you always get a hit, it would remove the need to do this as a distributed project, that would be the equivalent of having a radar device that can locate the diamonds directly.

In reply to your second question, we are currently working in improving Rosetta more than trying to get useful work from it. This thread started by David Baker describes some of the things that are planned, but as of right now we're still working on getting Rosetta as good as it can be.

---

ah, so my results are actually useful?
are they trying to improve rosetta along the way, or are they just trying to get science results from it?

They are actually squeezing scientific results from your computer and improving the methods along the way!

---

thankyou both, i have a final question, is there a way/place to make suggestions about improvements to rosetta, or just state observations? I read in a few posts by david baker that he's keen to get feedback, especially with the new screensaver