Disqus - Latest Comments for mndoci

Re: Post Hadoop World thoughts

mndoci — Sat, 10 Oct 2009 16:15:07 -0000

Likewise. And hopefully under less busy conditions.

Re: Blog away

mndoci — Tue, 06 Oct 2009 23:39:27 -0000

I have no idea what I was thinking about at the time. My memory is not that good :).

I think what I was speaking about blogs as a whole, rather than specifics. I do agree that in the past year the quality of blogging has ratcheted up

Re: Trapped in the USA: Bioinformatics in the Near Concurrent Future

mndoci — Fri, 04 Sep 2009 01:16:06 -0000

Yep. Saw the talk at Supercomputing last year too

http://mndoci.com/2008/11/24/mapping-and-reduci...

Re: Trapped in the USA: Bioinformatics in the Near Concurrent Future

mndoci — Wed, 02 Sep 2009 02:13:32 -0000

Bosco, did you ever end up trying out Clojure? Might be an interesting approach to analyzing very large trajectories (driving Hadoop or some other map-reduce engine).

Re: Yes, Very few individual human genomes have been fully sequenced

mndoci — Wed, 26 Aug 2009 01:07:00 -0000

Presumably once the 1000 genomes project is complete then we have a 1000 :). Without saying who and how, I know of at least three more unpublished ones.

Re: The quirks that make drug development so hard

mndoci — Wed, 26 Aug 2009 00:50:11 -0000

Let me see if I can pull up any. Most of my knowledge is from conferences and being part of that world

Re: Putting the cart before the horse

mndoci — Tue, 25 Aug 2009 10:52:54 -0000

I'd love to know. It's something I'd like to explore given time. Functional programming has been around for a while, but somehow never made it into this realm. There are challenges to be sure. For one a lot of these problems are task parallel not data parallel. What you will see is people not using task parallel approaches to solve data parallel problems.

The other aspect is speed. The libraries, etc that have been developed in Fortran, C, C++ etc are important to big numerical computing.

As for big companies, if any will be successful here, it will be Intel, IBM, HP, since they understand many of these problems more than any, at least on the programming model side

Re: Is XML bad for big data?

mndoci — Sun, 23 Aug 2009 13:19:34 -0000

I am not quite as militant as Mike, since I do agree with you on that hammer and nails comment. However, in my experience (and all you have to do is look at CDISC, HL7, etc), what happens is people try and do exactly that and it slows down progress as you get locked down in committee. And I speak directly to the sequence part. In many of the transport protocols being developed, the sequences would NOT be transferred as sequences, but encapsulated in XML. That's the kind of bureacracy that Mike speaks to.

Re: testing feed

mndoci — Sat, 22 Aug 2009 02:09:00 -0000

testing

Re: DNA Origami and lithography

mndoci — Thu, 20 Aug 2009 03:19:30 -0000

Ben,

Spent enough time with biomolecular electronics to be more than a little cynical :)

Re: FriendFeed, Facebook and scientific communities

mndoci — Wed, 12 Aug 2009 09:40:25 -0000

Ian

Wonderful comment. I am not worried at all. I see opportunities everywhere and this, if approached the right way is certainly one. I am not leaving Friendfeed, not unless it decays or goes into directions totally anathema to my interests, and I am half as cynical as most. At the same time, I do believe we have enough people out there who can build stuff and it would be an interesting proof of concept, of an idea that's been in my head for a couple of years. We'll see. If we do go that route, should be fun.

That continuous, iterative development models is one of the lessons I take to heart. It's what has made FF so good. Always iterating.

Re: The Canonical model for scientific software

mndoci — Thu, 23 Jul 2009 10:04:23 -0000

I think one way to improve the quality is just adopting good software development practices. In the cases that work, like R, a strong core team, and appropriate practices ensure that is the case. On the other hand, in a lot of cases, you get sphagetti code, or the main driving postdoc moves on, and you get left with pretty darn bad code.

Commercial support is not a requirement, but it helps, not necessarily commercial, but an organization that identifies key challenges and weaknesses, perhaps the kinds of thing an academic group is just not incented enough to do. Obviously, linux and a scientific software package are a different beast, but this approach beats the traditional, "lets just in-license" the software and package it up and sell it.

Re: Hundred nanoseconds a day

mndoci — Wed, 17 Jun 2009 10:06:59 -0000

John

Completely agree. The physics could be improved greatly. Simulations today do tend to stay together, cause you are using explicit waters and better implicit models, but in general there are three challenges; compute power, force fields and programming models. We've got #1 in good shape, and #3 is better than before. #2 we need to work on.

Re: Data produced, analyzed and consumed. The impact of big science

mndoci — Wed, 01 Apr 2009 19:35:56 -0000

Hari

Doing a lot of what it takes to get big data production done is grunt work. It's industrial strength production. The MUCK as we would call it in these parts. Small labs should not be spending their time learning how to spend production systems that need to be optimized for cost and speed. It's just a waste of resources. The consortium idea doesn't work either. Too many egos, not enough efficiencies.

Re: Still waiting for that biosciences startup school

mndoci — Wed, 01 Apr 2009 15:31:13 -0000

Whoops!!! Fixed

Re: Open Companies? Not so fast

mndoci — Sat, 28 Mar 2009 20:07:19 -0000

I would love to see a concept like this pan out, but any success will almost certainly be an edge case. I do like the idea of being in a situation where you can jump from project to project, getting them of the ground, maybe returning every now and then.

Re: Industry watching: Biotech finally makes a profit

mndoci — Sun, 22 Mar 2009 00:28:11 -0000

Also, at least in my experience, many scientists/professors think their ideas are worth more than they really are. Another think I have a problem with is that in many cases the people who have invested the most time and energy into various discoveries, i.e. grad students and postdocs, don't often get to run with those projects cause of the way the whole tech transfer process is set up. I'd actually love to hear more from people on this particular subject (University tech transfers)

Re: You are not alone

mndoci — Mon, 09 Mar 2009 19:39:37 -0000

Emulation is one thing, creating or coming up with the idea is another. Science is governed by fact. Art isn't. There is a science to art and music, but their is a good deal of interpretation and subjectivity. In the end, once we have a full understanding, there is no interpretation, no assumption. If you gave someone not named Michelangelo the change to paint the Sistine Chapel, it would not be the same. The laws of motion though, are unique. Therein lives the difference.

Re: Pushing the boundaries of web publishing

mndoci — Mon, 09 Mar 2009 17:20:10 -0000

Just one reason I so dislike PDF as a web publishing format. It should be an option, but most journals treat web publishing as a way of disseminating PDF's. I wonder what impact services like Mendeley might have in potentially helping address this issue

Re: Sage: Data from old stomping grounds

mndoci — Fri, 06 Mar 2009 18:26:10 -0000

What you and I think is what we think it is. Amazing!!!

That article pretty much spilled the beans. Had a longer discussion with someone on this today. Definitely lots of people paying attention

Re: Bio-Mirrors: To be, or not to be

mndoci — Sun, 01 Mar 2009 23:39:53 -0000

Abhishek,

As mentioned, the UniProt architecture is a class "no single point of failure" design, and serves to provide fast response to requests.

The life sciences can learn a lot from the web in general where people are serving up information globally. Different use cases, but many lessons there

Re: Sage: Data from old stomping grounds

mndoci — Sat, 28 Feb 2009 04:10:03 -0000

I think I fixed it. Had some leading blank spaces in the API key

Re: Sage: Data from old stomping grounds

mndoci — Sat, 28 Feb 2009 04:08:06 -0000

Sooooo ... is it what I think it is :)?

Let me check that Facebook thing. I believe you're the first person who's tried it

Re: Crunch that data

mndoci — Sat, 28 Feb 2009 01:26:33 -0000

Dhruv,

That's great. Hopefully some of the folks that are in The Biogang will get do some forking and get machetEC2 bio friendly.

Re: More Singularity madness

mndoci — Tue, 17 Feb 2009 13:59:16 -0000

Let's have that discussion when we figure out how to fold a small protein or simulate protein transport properly. We are so far away from that, and there are a million things to do before we can even think about solving.