Sunday, March 22, 2009

Prima Cala


This beach is near to my home (a 5 minutes walk). I come here whenever I need a little of fresh hair. The place is often windy, like today.

Friday, March 20, 2009

Equinox

Thursday, March 19, 2009

Spring

Friday, March 13, 2009

SDBS beautified

The Spectral Database for Organic Compounds is very popular
and deserves all my praise: is straightforward to use, easy to access, rich of data, void of advertisement and completely free. Everybody has used it at least once, what else should I say?
Jaume Farràs Soler sent me a shell script to import a list of peaks, copied from SDBS, into iNMR. He said that the pictures of SDBS were not clear enough, while in this way he could generate much better ones. The address of this script is now:
http://www.inmr.net/library/sdbs2inmr
Later on we'll see why it can be useful. Let's start with a more immediate approach. First thing first: if you have a Mac, download the latest version (3.1.4) of iNMR passion (freeware). Now, the next time you find a 1-H spectrum on SDBS, scroll down the page, until you find this button:Clicking the button leads you to a list like:

  Hz    ppm   Int.

1367.55 3.422 485
1360.72 3.405 942
1353.88 3.388 515
 755.00 1.890 118
.......................


Copy the list, including the header: "Hz ppm Int.", into the clipboard. Open iNMR passion and choose, from the menu, "Simulate List of Peaks". You will instantly see the spectrum, synthesized, as large as your monitor, ready to be explored (or printed) and, it goes without saying, anti-aliased.

Caveat: I guess that SDBS reports the apparent eight of the peaks, which is given by the proper height of a peak plus the contribute of any overlapping peaks. iNMR interprets this value, instead, as the natural height. The result is that all the peaks of a multiplet will become taller, while the singlets will be OK.
The areas will be a little distorted. What's also missing from the SDBS list is the width of each peak. In the synthetic spectrum all lines appear equal and narrow (1 Hz wide). The areas of broad peaks will be strongly underestimated. Add to this the effect above (for overlapping peaks).

If you want something more realistic, you can use the script by Jaume. It will generate a list with widths. Import the list into the deconvolution module of iNMR, where you'll be able to modify individual line-widths and intensities. The spectrum will look more realistic, yet don't ask me where to get the correct values from.

If, instead, you know about another database, with a different list format, send me the address. I'll write another version of iNMR passion that will recognize your format too.

Tuesday, March 10, 2009

Evgeny wants you

Last year he opened the NMR wiki. This month Evgeny Fadeev has started a discussion group. Synthetic description: "All about magnetic resonance, summaries go to NMR Wiki". Evgeny also says: "Send invitations to your friends and ask them to invite their friends. Then we'll make it work!!!".
You are in time to become one of the first 100 members.

Sunday, March 08, 2009

Making the Headlines

The NMR discovery of the month, according to spectroscopynow, is the whitening method. When this humble blog introduced the method in October, I didn't believe it was SO important. I am still pinching myself.

Split Plot Splats

Friday, March 06, 2009

ini386

The reader ini386 compared the ease of use of iNMR with that of the Adobe reader. Such a comparison is impossible. Both products and their histories are under the eyes of every one. You should know them! Today I am trying to expose a few facts that are never been secret, they have been just ignored. In other words: nobody really cares about what I am going to write.
iNMR has arrived to version 3.1.1, with 3.1.3 ready to appear next week, through an incredible number of changes and upgrades. In the beginning, the process could be explained in two ways: either the preliminary versions were not satisfactory or the users were impossible to satisfy (does it make any difference?). Afterwards the process became a little different. Once you know that it's so easy to release a new version, it's impossible to stop. You are happy with the program not because it's perfect, but because you know that any defect you find can be eliminated.
While there is a single author, iNMR is the product of many minds. Two hands only; many minds. The clear separation of roles is a winning strategy. The users are in command: their desiderata are normally satisfied. The exception is when they ask another application or a companion application; these requests are difficult to satisfy. If, instead, they ask for a new functionality or a single modification, they are the kings. The user, however, don't the internal mechanism of the program and can't imagine them. It was relatively easy in the case of SwaN-MR, where the data structures were the same (in RAM, on disc and into the dialogs). It's impossible in the case of iNMR: the data structures (which contain the same data) changes at each level and there's a great work of translation under the surface. For this reason the user/king can't tell to the servant how to do his job.
When it's working time, the servant is the king. He can decide, for example, to spend a week to finish a job that could have been done in a couple of hours. The fun is in trying all the possible variations (and not having a deadline). Misunderstandings are frequent (and funny too): the user asks for a certain function, the programmer understands a different thing. Eventually both things are done (and remain; and the program grows...).
The opposite situation (democracy) is dull and stressing. In this case there are the meetings, boring and time-consuming, where there's always people who need to speak even when they have no (useful) idea to communicate. Often you can also encounter their historical enemies, and the psychodrama is on the show. Now, if you know that a new proposal for a change or for an experiment must be approved by a meeting of this kind, you'll be scared of proposing anything (and you'll never desire to experiment).
Don't get me wrong. I know very well that the concentration of resources (energies, capital, skills, everything) is the road to success. I am convinced that I go nowhere if I am alone. What's difficult (and probably never required) is to find an agreement between the coworkers. That's impossible while you are inventing something new. When a problem has already been studied very well, when several books have already been dedicated to the subject, then it becomes possible to work as a team.
Programming is becoming more and more similar to chemistry. When intermediate structures become important and popular, they are prepared commercially. Teams of programmers work together to create general-purpose libraries. Individual programmers can assemble the ready-made pieces to create the final applications. When the closest intermediate can't be found, we can still copy (recycle) the ideas, as we do in chemistry.
I couldn't have written this post 20 years ago. I was a different man. I used to say to the users: "This is the program. Learn it." And they kept saying: "Have you ever though about adding...? Why don't you imitate that other program...?". We have kept saying the same things over and over until today (and tomorrow). What is changed is that now I pay more attention of what they say, while they understand more of what I am saying. In the past I was happy if I could change their minds. Today I am happy when they have changed mine. It's not a virtue, it's my beard that's becoming grey. When I had a whole life in front of me, I tried to make the world a better place where to spend this apparently unlimited existence. I tried to make people think just like me. Today I enjoy to learn new things, to embrace new ideas. It's not as having another chance in life, it doesn't even come close, but it has a flavor of it.