Heya @GeoffHutchison this was a pretty nice article :) phys.org/news/…

congratulations on the grant!

@GeoffHutchison My biggest concern would be quality assurance. Is there anything in place to make sure thing being deposited is accurate/correct?

I think if the new users want to ask question from their homework, just let it be. They just want their question to be answer.

hello

Hi

dissociation (en.wikipedia.org/wiki/Dissociation_(chemistry)) requires energy, right? where does this energy come from?

@Simon-Nail-It We have a homework policy in place that allows for users to post HW questions. If the policy is not adhered to, the post will be closed or put on hold (as with any other post that doesnt meet this site's criteria).

At the end of the day @Simon-Nail-It, we are NOT a homework service.

@LordStryker We can message them to try to find the answer to the question, and edit the question by putting their own answer. After that, they can ask that whether the answer is correct or not.

@Simon-Nail-It we have a homework policy. If it isn't followed the post is closed.

@Simon-Nail-It if you would like to see the policy changed, I encourage you to post in Meta with your ideas.

@mrpyo I guess when sugar is into its simpler form, it will naturally react to oxygen to produce energy

@LordStryker Nothing much to say .

@Simon-Nail-It Don't expect anything to change then. :\

@LordStryker just a slightly change

@LordStryker Just let me think about this then maybe post some ideas on Meta

what software can I use to draw ball and stick models?

@JoeStavitsky ChemCraft, Molden, ChemDraw, Avogadro, Molekel, GaussView, etc.

@JoeStavitsky avogadro.cc/wiki/Main_Page One of our users here is an Avagadro developer (@GeoffHutchison)

So i'll go ahead and put a plug in for that. Plus, its open source.

@LordStryker which would you say is simplest? I'm in freshman chem and I just need to do a report

I think Avogadro would be given that its free and works on the windows platform.

Chemcraft is easy but you have to pay for that (its got a trial version though)

Gaussview cost $$. Molden is not easy to get running on a windows machine for a beginner. Same with molekel (i think).

I'd try out Avogadro first.

@LordStryker: avolgadro is awesome ty

@JoeStavitsky You should tell that to @GeoffHutchison

@GeoffHutchison thanks for an awesome program :)

@JoeStavitsky Thanks. Definitely, it's designed to be a "molecular sketchpad" for 3D views of molecules. That's why we created it.

Now we're working on v2, albeit a bit slower than I'd like, with much better and faster rendering.

@LordStryker Our quality assurance is that we're not taking deposits. We run all the calculations.

So people will be able to add molecules to the queue, but the calculations will only run every week or two to minimize "please do my homework" efforts.

I've learned from participating on Chemistry.SE. :-)

@GeoffHutchison How many jobs can the queue handle? Can this be done via an SSH interface or will it be a webpage-based one?

I could load you up some molecules to compute ;)

@LordStryker At the moment, we're starting with ~100,000 calculations at the PM7 level, then B3LYP/6-31G* and wB97-xD

How will you treat convergence issues?

The submission will be via web, but if you have some interesting structures, you can e-mail me or send me a URL. Right now, we've grabbed coordinates from PubChem, etc. so we're short on metals and non-drug-like.

We have a set of scripts to check if a job has run successfully.

Basically, I know some things will require manual intervention, so a sizable amount of $$ will pay work-study students to check failed jobs.

My feeling is that we can pay better and offer much more interesting work than the dining hall, so we'll give preference to diversity and financial need.

I have the backend in pretty good shape, I think. (Famous last words.)

The web front end.. Well, I know zero about mobile "responsive" web design.

And we'll definitely have a "report bug" link, which I think was on the mockup iPad graphic that circulated with the press release.

I think your idea is pretty amazing. I'm looking forward to seeing how this pans out. ETA on launch?

We've been doing some internal testing now and hope to have a beta website up sometime in the summer, if not earlier.

We've got the PM7 calculations done (300 compounds per hour) so it's a matter of getting the website ready to distribute them.

I suspect when we put the code up on GitHub, we'll get lots of feature requests. :-)

And of course I still owe Chemistry.SE some pretty pictures.

Yeah, indexing/cataloging on the fly sounds complicated!

But hey man, congrats on the grant. You're doing good work.

Oh, we're not going to index / catalog on the fly. That's why we're using a queue and updating in batches. :-)

and just a heads up, I can almost guarantee @Martin will want a word with you regarding B3LYP

Haha. Yes. We'll do it simply because people will ask if we don't.

"Why don't you use B3..." SMACK Because it sucks!

;)

Problem solved

One big thing we're doing is to have a page that discusses the "error bars" for different properties for the different methods.

This will cut down on the complaints of "your dipole moment for molecule X is wrong."

heh

But it will also let people consider different methods, and allow us to teach about that in classes.

What prog. are you using to do these calcs?

MOPAC for MM?

Spartan has been selling this as part of their package, but it's expensive. We're using unused cycles on our campus cluster.

MOPAC will be used for PM7, obviously.

We will likely use Q-Chem for other jobs.

Gaussian expressly forbid this concept, since in principal, people would buy fewer licenses.

Hey, get banned. Display letter as trophy. Profit.

I love QChem.

That documentation is just... phenomenal.

Pitt will eventually be banned, since my colleague is a Q-Chem developer. But we need to convince some of my colleagues that life will be fine (and probably better).

Yes, the docs are great, although I'd love it to be in HTML in sections.

@GeoffHutchison Gaussian doesn't ban universities though anymore...

wouldn't the ban only apply to your colleague?

No, it applies to anyone who collaborates with him, so .. well most of the department.

bannedbygaussian.com here Geoff comes!

Well hopefully it won't happen for a long while.

(shrug)

I was banned in grad school, since John Pople was at Northwestern.

The main downside would be that Open Babel and Avogadro might not easily keep up with changes to file formats in G15 or whatever the next version will be.