- Shape-univ - cosmo.torun.pl

w_Q \approx -1.5 ??? (DE-V0.04)
by Boud Roukema 02 Jun '02

02 Jun '02

Cze�� wszystkim, I've done a bit more documentation. Now you should be able to download, compile and run DE-V0.04: http://www.astro.uni.torun.pl/~boud/DE/ and get some nice correlation function plots averaged from all six high completeness 2QZ fields, very similar to the ones in RMB2002. They're not exactly the same, because (i) the random number seeds are almost certainly different and (ii) there are probably small changes like different numerical accuracy on different computers, and small changes made to the programs since the time the results of the paper were obtained. Guess what? I couldn't wait to work on the next step, even when the documentation for the earlier steps is not finished... So I wrote DEdNdzfull (in DEscramble.f) which instead of using (scrambling) the z distribution of just the individual field being analysed, uses the full z distribution of the 2QZ-10K catalogue. It seems to me that this is a better way of producing the "true" redshift selection effects for this particular catalogue than the idea I had before of passing a "typical" QSO spectrum through the optical u, b_J and r filters. The z-scrambling technique means that some *real* signal may have been cancelled out, so the results were conservative. Now, with a more "neutral" z distribution for the "randoms", the signal should in principle be stronger - if it is real. Well, the resulting signal is spectacular! :) :) :) :) We still have a lot of work to do (I have to get plot_cf working), but it seems clear that (0.3,0.7,-1.0) gives a bad fit, and that (0.3,0.7,w_Q) with w_Q \approx -1.5 is likely to give a better fit and signal, which would greatly upset all the theorists! Anyway... during today's session you should hopefully be able to calculate these functions and see for yourself... Na popo�udnie o 14:00 godz, boud

2 2

Re: comments from Charlie Lineweaver
by Boud Roukema 31 May '02

31 May '02

Cze�� wszystkim, Gary is giving talks on RMB02 in Australia on Monday and Tuesday: On Fri, 31 May 2002, Gary Mamon wrote: > Hi Boud, > > With the better z-distribution, the 122 peak is back. With a vengeance. ;) > > Ah? Which better z-distribution? Has Brian Boyle and colleagues released > a larger sample of quasars now? If you have any new plots on this, please > send them at least by Monday morning your time, so that I can show them > for my Mt Stromlo talk (the Macquarie one will probably be less well > attended). Gary, you should be able to download, compile and run DE-V0.04 http://www.astro.uni.torun.pl/~boud/DE/ The format is somewhat similar to ArFus - there are html help files etc. Macquarie: Mon 3 Jun http://www.physics.mq.edu.au/talks/index.html (Some wrote "10QZ" instead of "2QZ-10K" ;) Stromlo: Tue 4 Jun http://msowww.anu.edu.au/astronomy/ozcalendar.shtml shape-univ course: Qn 1 I think we should discuss this afternoon whether we are ready to show some of our new results at these two talks! (It's the old question of what level of "paranoia" - about others taking our results before they are published - is reasonable.) IMHO, I think that Gary should show the new results. Qn 2 Should Gary invite people at the lectures to download DE-V0.05 (which should be ready by Monday) and have them play with it? MSZ, yes, but I'd like to know what everyone thinks... See you in 1/2 an hour, Boud

1 0

cosmology textbooks
by Boud Roukema 24 May '02

24 May '02

<cross-post shape-univ, cosmo-torun, cfalib> Cze�� wszystkim, Printed books are expensive and slow to get, but it *is* more fun to read a book than to learn everything online. So as more people get interested in cosmology at the TCfA it would probably be good to have some more copies of the 1990s generation of basic cosmology texts; and, say, half-a-dozen copies of Numerical Recipes (unless someone knows of a better book than NumRec, since there *are* some criticisms of the subroutines which do "complicated" things like differential equations etc., but in my experience the simplest routines seem to be perfectly OK). Karolina: would you please be able to find out * the cost and approximate delay in shipping for the following: Peebles 1993 Liddle Peacock, John A. Cosmological Physics Padmanabhan Structure formation in the Universe Padmanabhan Cosmology and Astrophysics - through Problems Padmanabhan Course of Theoretical Astrophysics - Volume III: Galaxies and Cosmology (in press) Numerical Recipes * It would also be good to find out what po polsku translations of cosmology introductory texts exist and how much they cost (I guess most are from Pr�szy�ski, translated by CAMK-Warszawa people ;) Everyone: to me it makes most sense to have the library manage copies of these (though I know that I'd like to have some personal copies). I'm not sure how we should decide how many copies to get, and of which we should get copies. (Except for Numerical Recipes: I think that 3-4 copies would be reasonable.) The advantage of getting several copies of the same book is that different students would be able to work with the same language, symbol, etc. conventions and content; but the advantage of having just one or two copies of each of several *different* books is that the material is somewhat different (some of the above say nothing at all about the topology of the Universe; a few may say a little...) and so the collective knowledge of students will be better than if, e.g., just one textbook is available. I think that part of the decision should come from the experience of students - how useful do they find the books, which have the English which is easiest to understand, which do they prefer generally? But it should also come from people as they start doing *research* in cosmology - which texts are the more useful ones? There's no hurry, but these are things to think about. Pozdrawiam, Boud ______________________________________________________________________ Details with URLs: Just start from http://www.astro.uni.torun.pl/ and go to library and you'll find the catalogue search here: http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl Peebles 1993 - 1 copy po angielsku TCfA http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl?kl=4467-1022250705&pp=… Liddle - 1 copy po angielsku TCfA 2 copies po polsku - 1 TCfA, 1 biblio g��wna UMK http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl?kl=4308-1022250625&pp=… Peacock, John A. "Cosmological physics" - 1 copy po angielsku TCfA http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl?kl=4553-1022250731&dis… Padmanabhan, Thanu no copy at UMK! http://www.iucaa.ernet.in/~paddy/books/books.htm Karolina, if there's difficulty getting the info, you could email Paddy at his admin address: <nabhan at iucaa.ernet.in> Silk, Joseph no copy at UMK! Karolina: could you find out the title(s) of Joe's recent general books on cosmology? Chances are he's probably written a dozen in the last five years ;), and I know for sure there's at least one with a title like "Cosmology". Numerical Recipes 1 copy in TCfA http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl?kl=5345-1022251291&pp=… 2 copies in Chemistry http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl?kl=5345-1022251291&dis… http://lx.bu.uni.torun.pl/cgi-bin/BG/Z2Wumk/z2w_f.pl?kl=5345-1022251291&dis… ______________________________________________________________________

1 0

DE-V0.03 released (just in time ;)
by Boud Roukema 24 May '02

24 May '02

Cze�� wszystkim, Just in time for me to catch the bus to Piwnice, voil� version DE-V0.03, with more documentation (i.e. all the routines in DEreadVeron.f + the z scrambling routine, DEscramble.f) http://www.astro.uni.torun.pl/~boud/DE/ The flowcharts now show the *test* routines in green. A good test of the quality (usefulness etc) of the documentation would be to see if you can make some simple modifications of the test routines, and recompile and run them and see if the routines they test seem to work properly... See you in two hours, Boud

1 0

-lX11
by Boud Roukema 22 May '02

22 May '02

Micha�, there are comments regarding "-lX11" in the makemake file in DE-V0.02. Do you think these are sufficient, or should we put them somewhere else? My idea is that the user should edit the makemake file, and not the makefile itself, for things like PGPLOT environment variables, libraries like X11, etc. Boud

3 9

-lX11 (fwd)
by Boud Roukema 20 May '02

20 May '02

My first copy of this seems to have gone missing. Is the sympa daemon down? Or is putting a "-" at the beginning of the Subject: header a problem? > ---------- Forwarded message ---------- > Date: Mon, 20 May 2002 14:36:39 +0200 (MET DST) > From: Boud Roukema <boud@... > To: shape-univ@... > Subject: -lX11 > > Micha�, there are comments regarding "-lX11" in the makemake file > in DE-V0.02. > > Do you think these are sufficient, or should we put them somewhere > else? My idea is that the user should edit the makemake file, > and not the makefile itself, for things like PGPLOT environment > variables, libraries like X11, etc. > > Boud

1 0

DE-V0.02 more or less OK? :)
by Boud Roukema 20 May '02

20 May '02

Cze�� wszystkim, * report on Fri 17 May lecture * credit for course * future * long term future (wrzesie�/listopad 2002) * short term future (czerwiec/lipiec/sierpie� 2002) ------------------------------------------------------------ * Fri 17 May lecture * Welcome back after 3 maja & juvenalia holidays! On Friday 17 maja the people who turned up had already downloaded DE-V0.02, and tried to compile and run. I helped with some minor comments, and everyone succeeded in getting things to the stage of running DEplates for one "field" of the 2QZ data, for one choice of Omega_m, Omega_Lambda, and plotting a real correlation function from this using plotcorrn. Of course, some more work is still needed before getting correlation functions as nice as in the final results... We discussed how to go from here, and it seems that within the constraint that people just work on the project during the Friday afternoon session, we have an efficient working style which is: - Boud adds documentation and corrects bugs during the week - on Friday, everyone else compiles/runs the new version and gives feedback on what bugs there are which seems fine to me. Of course, (see below), there are only two official lectures left... * credit for course * I think I need to tell Andrzej Strobel/Woszczyk which students have officially participated in the course, so that you get credit for it. This is the list of students based on the list membership (excluding PhD students and faculty, who don't need course credits!), please tell me if you think there is anyone missing who should get a credit: - Micha� Fr�ckowiak - Marcin Hajduk - Rafa� Kosi�ski - Hubert Krzemie� - Bartosz Lew - Pawel Ludian * future * There are only two Fridays left in the academic year (if I've understood the calendar correctly): Fri 24 May 2002 Fri 31 May 2002 * long term future (wrzesie�/listopad 2002) * I will ask Andrzeji Kus/Strobel/Woszczyk what formal possibilities there are for a course next semester, but it seems to me that what is important is: - that there is a "lecture course" where students (and faculty) can learn and get their hands on doing real cosmology research - that the "undergraduate" students (I-V roku) can get credit for the course, since it takes time away from their other coursework - that it would be best without exams, but if there is an exam, that the students are equally involved in designing the exam with the lecturer Any opinions? * short term future (czerwiec/lipiec/sierpie� 2002) * I should probably take some holidays and go to some conferences, but so far my plans are still vague. If anyone (including III roku students) thinks that he may wish to continue on the project during czerwiec/lipiec/sierpie� 2002 please say it on the list, or else at the next Friday lecture, as it is more interesting to work *with* someone than alone - maybe several people wish to work together at the same time. Probably it would be best if you put a probability. Here are my probabilities that I will be available in Toru� to work on the DE project in the next few months: czerwiec 2002 95% lipiec 2002 80% sierpie� 2002 50% If several people will be here in June (czerwiec), we could continue with the same time on Fridays, if this is convenient. Na ra�e Boud

1 0

DE-V0.01 compilable + runnable -> antipode ?
by Boud Roukema 13 May '02

13 May '02

OK everyone, Thanks to your comments, DE-V0.01 has now been released! There's not yet any real documentation on the program modules themselves, but what I've done is documentation on how to manage the whole package. Compiling different routines and linking them together is generally done with a makefile, but different versions of linux/unix use different syntax or levels of generality, so I've followed the pgplot solution which is to have a shellscript file makemake which makes the makefile depending on what system you're on. Well, if you download the new package (either with or without the data files; "nodat" means without the data files) from: http://www.astro.uni.torun.pl/~boud/DE_tar/ and then read through the web pages: http://www.astro.uni.torun.pl/~boud/DE/ you should be able to make the library (libDE.a) and compile a program like antipode and see at what redshift the antipodes are. In fact, if you just type make antipode after you've made the makefile, then libDE.a should be made for you automatically... Next on my list: - documentation within each of the more important routines - cosmological routines - routines for reading data - main programs which do the work Cze�� Boud

1 1

off-topic: taxing free software - total nonsence!!!!!!
by Michal Frackowiak 26 Apr '02

26 Apr '02

Polish taxing institutions were always hungry for taxing free software. But so far they only tried to tax USERS. Now they are preparing to tax PROGRAMMERS!!!!!!!!!!!!!!!!!!!!!!!!!!! So far the new law gives the chance to tax programmers but the question is if the taxmen will be really so ignorant about the idea of free software? Unofficially the Minister of Finances is supporting the idea of taxing free software. But we shall wait..... The paradox is because the novelisation of tax law assumes giving a licence (selling) should be taxed. And that is wise. Free programmes also give licences (mainly GPL). On the other hand if a given service is done free of charge when one usually get paid for it, this service should be taxed. => giving a FREE licence is not common, so it is an exception, because licences are always SOLD. free programers give for free what is usuallu sold. => they should be taxed It is sad. No more comments. Michal F. http://www.gazetaprawna.pl/numery/05602-661/aktualnosci.html?id_gpol2=3554

1 0

Intro to cosmology (my master thesis) %LOOK%
by Michal Frackowiak 26 Apr '02

26 Apr '02

Hej! Ponizej posylam link do szkicow mojej pracy magazyniersiej. Mam nadzieje, ze moze to posluzyc do "lagodnego" wprowadzenia do kosmologii. To, co sie tam znajduje w tej chwili, to: * wstep (historyczny?) i kilka slow o kosmologii, problem stalej kosmologcznej * podstawy teoretyczne: r-nia Einsteina, dodatkowe pole kwintesencji - z metod wariacyjnych, wyjasienia podstawowych pojec w kosmologii * troche szczegolow o samej kwintesencji * i wiele wiecej, ale to pozniej (supernowe Ia, anizotropia CMB) prosze nie zwazac na literowki i bledy, tekst jest napisany, ale go jeszcze nie czytalem ;) mam nadziej, ze bedzie uzyteczny pozdrawiam Michal http://www.ncac.torun.pl/~michalf/MasterThesis/

1 0

DEflow1.eps printable version
by Boud Roukema 26 Apr '02

26 Apr '02

Cze�� The flowchart I sent yesterday was written with xfig There is a bug in xfig when writing a postscript file - it does not write the final end statement showpage So you can see the file with ghostview or gv, but if you print it nothing will happen - sorry. The solution to this bug is to edit the postscript file, and add the line showpage as the last line. Then the file is printable. Here is a printable version of DEflow1.eps. Boud

1 0

rough flowchart
by Boud Roukema 25 Apr '02

25 Apr '02

Cze�� wszystkim, I've prepared a flowchart and realised that only *one* of the "cosmological" function/routine files - pdWeiq.f - is called by plates, (pdWeiq or pdWein is called for getting radial proper distances), but angle4D8 (in angle4D) which I've called "numerical" is also called and is actually very useful conceptually - it's what is used to get the distance separation between two points in arbitrarily curved space. [Just as a "distance" on a 2-sphere is an arc-length on a 2-sphere embedded in R^3, a "distance" on a hypersphere or 3-hyperboloid is an "arc-length" on a 3-sphere or 3-hyperboloid embbeded in R^4 or M^4 (Minkowski-4).] * the flowchart will be correct once I modify the names to add "DE" and add documentation, and I haven't yet written DErcurvq (which should be like rcurvfn in pdWein.f) * the arrows show function calls. Na jutro pi�tek o 14:00! Boud

1 0

documentation; What name for our package?
by Boud Roukema 17 Apr '02

17 Apr '02

Cze�� wszystkim, I am very glad that several of you (at least three people) have had the courage to tell me that I have to do some work (in documenting the software), otherwise we will not get anywhere in our "monograph course"! This a good sign for learning how to work together :). So, I'm about to start preparing some documentation. My plan is to follow the PGPLOT style documentation which was used for the ArFus galaxy formation package: http://darc3.obspm.fr/~roukema/ArFus/index.html (You can download a *local* copy of the full package, including small example N-body simulations here - total size 31Mb: http://www.astro.uni.torun.pl/~boud/ArFus_tar/ArFus-V0.03.tar.gz ) Question: what name do we want to call our package? I think the name should not be too long, and it would be good to have a two-letter abbreviation to put in the subroutine (module) names, like "PG" for pgplot, "AF" for ArFus. Some ideas with comments + for - against ... ShapeUniv - SU + easy to say - confusion with SU groups of superstring theory? - confusion with Soviet Union? DarkEnergy - DE + easy to say - confusion with the name of a neighbouring country? FormaWszech�wiata - FW - confusion with "forward" in emails - difficult to say "FW..." for the name of a subroutine - includes a non-latin-1 character ObservCosm - OC - pretentious (claims to cover all of observational cosmology) TorunCosm - TC + maybe the best? - confusion with TCP/IP? QuintEssence - QE - confusion with Queen Elizabeth? CosmParameters - CP --- confusion with Communist Party KoperNik - KN - too many things are called Copernicus in Toru� & Poland - pretentious? Well, I've spent enough time on this, I'm sure others have some ideas. I'll put "DE" for the moment - we can change this later with a shell script depending on what people choose. Na ra� Boud

2 1

summary 14/04/2002
by Boud Roukema 12 Apr '02

12 Apr '02

Cze�� wszystkim, In today's lecture we regrouped, the two groups were Rafa�/Jacek/Hubert/Micha�-F Marcin/Pawe�/Boud In my group, Marcin and Pawe� looked through some of the subroutines in cosm.tar.gz and nrec.tar.gz and I tried to explain what the important ones did, and in the other group people tried compiling. * documentation It is clear that I have to write some good documentation, and quickly, so that everybody is not slowed down by illegible programs... Rafa�'s suggestion for documentation: - what the program/subroutine does - what external program calls there are - what data it uses * reference books Other discussion was about reference books for modern cosmology, some of the good recent books are: - Peebles - Liddle - Peacock Maybe someone could find out if we have these in our library, or who has borrowed the copy(ies) of these. Another reference book we should have is Numerical Recipes. Probably later when we have some grant money, we could follow IUCAA/IAP type policy and have something like 2-3 copies of each of these key references. (But this also requires trusting each other - the books are expensive and need to be shared around and kept track of by the library!) Of course, online references cost nothing, though of course it's nice to have printed books. I think online sources are probably the best for correlation functions and w_q. * computer access Andrzej M - could you coordinate with Hupert to tell him about what modem access we have, if any? He lives in a small town 30km from Toru� and apparently has no local facilities except his own PC. OK, watch this space for documentation, let's see if I can get organised... Pozdrawiam boud

1 0

... we're in pracownia computerowa w radio astronomii i w biurze bouda...
by Boud Roukema 12 Apr '02

12 Apr '02

1 0

14:00-15:00 cosm routines + antipodes; 15:00-15:30 feedback
by Boud Roukema 12 Apr '02

12 Apr '02

Cze�� wszystkim, Sorry for writing so late - it just goes to prove that my own organising skills need improving! Last week we more or less went through the points I had suggested. I thought that dividing into small groups with a mix of "generations" was best; Micha� F thought that people who already know each other was best. The final result was to divide people randomly, so here are the initial groups: Tomek/Micha�-F/Rafa� Radek/Hubert Marcin/Andrzej-M/Boud We have at least one person with a computer in an office at Piwnice, so the plan was to have the lecture as follows: * lecture structure Friday 14:00-15:00 each group meets around a computer and see what can be done for playing with the programmes, learning how they work and working towards the correlation function etc programmes 15:00-15:30 we meet back together in the lecture room, have coffee and discuss how things went, problems, what to do next, etc. (people who do not catch the bus can stay later) * comment: some people have more time, some have less time (e.g. only the Friday lecture time). I don't see this as a problem - different people should be able to accept complementary time usages. This was part of the exercise with getting to know one another... [English language note: "complement" .ne. "compliment"] * I still think this is a good plan - I got some feedback that Rafa� and Micha� could not compile anything of my programs ;), but below is some initial stuff which I think should be able to keep people occupied for 1 hour. If you can get through that, you'll have some idea of my programming style and how to decode it ;) or improve on it. Exercise: find out some local cosm parameter values for which we could see the Galaxy (or the spot where the Galaxy was going to form), for a trivial topology! ###################################################################### **** Today, 14:00 12 kwi�tna 2002 **** ----- I will wait at the lecture room for 10-15 minutes (14:00-14:15) for any new people that turn up, or people that were missing last week. E.g. Pawel, Bartek, Micha�-H, if you arrive I'll explain to you what we did last week and you can join the existing groups (or if there are too many people, we can start a new group). ---- ---- At 15:00 we meet back at the lecture room and give/receive feedback. ---- ###################################################################### See you in 20 minutes! Micha�, thanks for the C references! Boud To get people started, here is a rough guide to cosm.tar.gz ---------------------------------------------------------------------- cosm.tar.gz e.g. unload in a directory called cosm/ ## x makefile, 1681 bytes, 4 tape blocks - you'll need some routines in nrec.tar.gz, e.g. unload in a directory called nrec/ - you should read through the makefile in nrec.tar.gz, make some directories, and then type make nrec.a in the directory with the source and then make appropriate links back in the cosm/ directory ## x d_l3.f, 14623 bytes, 29 tape blocks ** proper motion distance ** - callable functions - calculates proper motion distance as a function of z - calculates inverse function: z as a function of proper motion distance - integrates an initial table the first time it's used - test program at beginning - depends on Om_m, Om_Lambda (called "cc_0"), w_q test compile: switch "program" and "subroutine" at top, and make d_l3 x d_l.f, 732 bytes, 2 tape blocks - like d_l3.f but no Lambda, w_q dependence x d_l2.f, 14135 bytes, 28 tape blocks - like d_l3.f but no w_q dependence ## x pdWeiq.f, 4968 bytes, 10 tape blocks ** proper distance ** - callable functions - calculates proper distance as a function of z - calculates inverse function: z as a function of proper distance - integrates an initial table the first time it's used - test program at beginning - depends on Om_m, Om_Lambda (called "cc_0"), w_q test compile: switch "program" and "subroutine" at top, and make pdWeiq x pdWein.f, 6835 bytes, 14 tape blocks - like pdWeiq, but no w_q dependence ## x tofz2.f, 6642 bytes, 13 tape blocks - callable functions: cosmological time as a function of z and inverse - no allowance for (w_q.ne.-1) test compile: switch "program" and "subroutine" at top, and make tofz2 ## x antipode.f, 16710 bytes, 33 tape blocks - This should be a fun test of your geometry intuition. Depending on values of Om_m, Om_Lambda, w_q, find out at what redshifts the "South Pole" and "North Pole" are (let's say that we live at the North Pole, after all, pogoda tutaj jest prawie blisko z pogodzem z Polem P�nocnej... ;) test compile: switch "program" and "subroutine" at top, and make antipode ## Exercise: find out the most "reasonable" Om_m, Om_Lambda, w_q values for which the "South Pole" etc. are at observable redshifts. ## Other stuff, probably less useful for our project. x cdm.f, 1910 bytes, 4 tape blocks x dVdz.f, 3559 bytes, 7 tape blocks x dVdz2.f, 4931 bytes, 10 tape blocks x schech.f, 1806 bytes, 4 tape blocks x units.f, 157 bytes, 1 tape blocks x zt.f, 9294 bytes, 19 tape blocks x zt_noL.f, 8746 bytes, 18 tape blocks ----------------------------------------------------------------------

1 0

programs + data
by Boud Roukema 12 Apr '02

12 Apr '02

Hi, Here are my existing programmes + data. There are certainly a few things like filenames and logical options which need to be set correctly for these to work - they are not (yet) user friendly. But you can see (more or less) what I've done. Boud

2 1

wyklad 5 kwietnia 2002
by Boud Roukema 04 Apr '02

04 Apr '02

Cze�� wszystkim, The following is my suggested agenda for the lecture this afternoon 14:00 Fri 5 Apr 2002 in the Radio building. (1) First things to play with :) ----------------------------- cosm.tar.gz nrec.tar.gz You'll find these attached - I haven't checked them, sorry, I'm tired and it's 01:06. (2) Poll (for finding out how to work together) -------- During the lecture I will ask everyone (including me!) to rate how he/she feels his/her level is from -5 (low) to +5 (high). The purpose is so that we each get to know where each other feels strong and/or weak, and we can help each other in their weak points. * self-confidence * organising skills * unix * fortran * C * pgplot * typing speed * comoving coordinates * curvature * topology * correlation function * null hypothesis rejection I will suggest splitting up into pairs/triplets and discussing these in separate groups for 20 minutes (that's less than 2 minutes per question, so you may not be able to discuss everything). Within each pair/triplet, please look at the strengths/weaknesses of the other 1 or 2 people and decide how you can concretely help the others in their weaknesses. In particular, for self-confidence, it would be good over the next week to make at least 1 compliment about the other 1 or 2 people in the pair/triplet. It should be a genuine compliment - and this will force you to try to really understand the person, otherwise it can't be genuine... (3) time scale -------------- 8 weeks till 31 May (4) possible programme ---------------------- a - play with cosm.a + nrec.a and document the routines in English & po polsku, see how much can be thrown away because redundant and/or useless and/or confusing b - get everything compiled including b1 -- reading in data files b2 -- calculating correlation functions b3 -- plotting correlation functions b4 -- plotting null hypothesis rejection levels c + add new routine for modelling ubr selection and avoiding z scrambling d + can calculation be speeded up? e + calculate for full 11K initial release f + write routine for reading in Las Campanas Distant Cluster Survey http://de.arXiv.org/abs/astro-ph/0201534 g + calculate result (should have degeneracy similar to SNe Ia degeneracy) h - write down method and results in English i ** positive/negative feedback loop ** Think about the results and based on thoughts, corrections, worries, iterate again through loop doing it better. j Keep iterating until convergence is achieved. kwiecie� 2002 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 maj 2002 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 (5) Question: ------------ Is it reasonable to ask everyone to write at least one email per week to the list saying where they've got to, what their problems, successes are? (6) Conceptual task: ------------------- getting null hypothesis rejection probabilities for the "D" test; how reliable, robust is the test? Useful reference? (at least for one of Boud's weaknesses, see (2) above ;) ) ----------------- http://www.le.ac.uk/cc/iss/tutorials/cprog/cccc.html Na popo�udniu o 14:00 (not 14:15)! Boud

1 0

holiday 29 march
by Boud Roukema 28 Mar '02

28 Mar '02

Noone has said anything about whether or not they want a lecture tomorrow 29 March, but since it is officially a University holiday, I'll assume by default that noone expects a lecture. But if anyone wants to come along and talk, feel free to do so! I'll be here (at Piwnice). Cze�� Boud

1 0

feedback 15 March lecture?
by Boud Roukema 18 Mar '02

18 Mar '02

Cze�� wszysce (1) starting time 14:00 proposed In order that people can leave at 15:35 to catch the 15:52 bus, Andrzej proposed to start the sessions at 14:00 instead of 14:15. Everybody present was in favour, I think. In case lunch arrives late at Piwnice, we could start the lecture informally in the dining room - I see no problem with this, it could help to get people relaxed and talking informally. So, could people please be ready to start at 14:00, in the interests of 15:52-bus-catching people? Thanks. (2) Fri 15 March lecture I'm not sure if we went through enough or too much, but I realised that although there is nothing *difficult* in http://de.arXiv.org/abs/astro-ph/0106135, there are *many* details to think about. Rafa� asked about (i) what needs to be done to my programs, (ii) why can't we just use them unchanged? (ii) If we use them unchanged, then we will just get the old results, instead of exciting new results. (i) Here's a rough guess at what needs to be done: - rewrite the programmes to be more modular + add documentation - add a new routine for reproducing the photometric u-b_J-r band selection effects, so that "z-scrambling" doesn't have to be used for the "random simulations" for correlation function calculations - check that everything works fine on a subset and then calculate for the full "10K" data set on a fast computer - write input routines for the filter match selected candidate clusters in the 0.5 < z < 1 (approx) range (I'll look up the reference...) - this should directly reproduce the Omega_m - Omega_Lambda degeneracy of the SNeIa data (3) correlation functions Rafa� also asked about correlation functions. I suggest the 0106135 paper for the full details, but for an easy introduction, the following paper is probably better: http://cdsads.u-strasbg.fr/cgi-bin/nph-bib_query?bibcode=1994A%26A...285..3… So for next week (14:00 pi�tek 22 marca) I'll try to prepare some information about my programs and the programs themselves... Pozdrawiam Boud

1 0

torus/klein bottle games :)
by Boud Roukema 12 Mar '02

12 Mar '02

Cze��, Rewording what I said on Friday, this is a lecture course where instead of exams, the lecturer asks you to play games. :) Witam, zagrajcie! http://humber.northnet.org/weeks/TorusGames/ For those absent on Fri 8 Mar: we did a review of curvature for new people, and then a quick introduction to topology. This Friday (15 marca) I propose to get back to curvature (or more strictly speaking, to the metric parameters, which include Omega_m, Omega_Lambda and... w_Q, the quintessence parameter) and to explain what Staszek, Gary and I did in: http://de.arXiv.org/abs/astro-ph/0106135 and what I think can be done better and with other public data sets. I think this should be enough to give everyone a feel for what needs to be and/or can be done, and I propose that the following Friday, 22 March, we make a decision on which shape-of-the-Universe project we wish to carry out, and start discussing how to organise this, conceptual questions, programming concepts, etc. Pozdrawiam Boud PS: Pami�tajcie: ta lista jest dla wszysce - wszystko mo�e pisa� na liscie!

1 0

particle horizon vs event horizon
by Boud Roukema 06 Mar '02

06 Mar '02

The simplest definition I found of the difference between the particle horizon and the event horizon is here: http://iapetus.phy.umist.ac.uk/Teaching/Cosmology/Metric.html The particle horizon is defined as the largest comoving distance from which light can have reached us - today. The event horizon is defined as the largest comoving distance from which light will ever reach us - at any time in the future. To have an event horizon, a universe needs to expand "very quickly", e.g. in a de Sitter universe, whose scale factor follows a pure (positive) exponential: http://nedwww.ipac.caltech.edu/level5/Watson/Watson6_1.html Apparently the definitions come from Rindler (1956, MNRAS, 116, 662): http://nedwww.ipac.caltech.edu/level5/Seitter/Seitter3_2_2.html but the figure is not clear to me. Boud

1 0

next lecture Fri 8 Mar
by Boud Roukema 04 Mar '02

04 Mar '02

Cze�� wszysce, On Friday 1 March we started from the beginning again, going through comoving coordinates and curvature, with an added touch of history. Where did Euclid spend 30 years studying geometry and writing an extremely good review paper? Cairo, Africa. That's where zero curvature geometry models come from. Western scientific culture is fundamentally African. [Anyone want to search for http refs?] Who developed and introduced algebra and the Hindi decimal system to the West? An Arabic-speaking Iranian scientist born in... Khiva, Uzbekistan, hence known as al-Khwarizmi, who spent a lot of his life in Baghdad: http://www.mscf.uky.edu/~carl/ma330/project2/al-khwa21.html So, the "gor" in algorithm comes from "Khiva" in Uzbekistan. http://www.ati-uzbekistan.com/english/uzbekistan/touristcentres/khiva.htm So Western scientific culture is equally Asian. (Do you really want to do physics using roman numerals? VIII + I = IX ?) Modern "Western" science involves a mix between thinking in terms of geometry (African), decimal numbers (Indian), algebra (Arabic culture) and algorithms/software (Arabic/Iranian/Uzbek), though of course there are some European and North/South American contributions too... IMHO, observational cosmologists tend to think in terms of numbers and software, theoretical cosmologists tend more to think in terms of algebra (in the more modern concepts of algebra, calculus etc.). We discussed whether people would prefer (1) continuing lectures in a similar more or less traditional style, or rather (2) move on to a research project involving some hands-on code-writing and observational data. ** standard ruler project ** The particular project ("standard ruler project") I propose is a rewrite of the code of http://de.arXiv.org/abs/astro-ph/0106135 and application of the same analysis to a larger superset and to different observational data sets, and using more refined observational corrections in order to get more signal out of the data. Anyone contributing will have their name on any publications resulting from the work (unless they choose not to!). This relates to the shape of the Universe in the sense that - Omega_m, - Omega_Lambda and - w_Quintessence can be thought of just as parameters of the metric which convert [right ascension, declination, redshift] into 3-dimensional comoving positions. If I understood people's comments right (please feel free to correct me), the initial preferences were: (1) Bart (2) Marcin, Rafa�, Micha�, Sebastian Since we didn't cover topology during Friday's session, what would seem to match both preferences is: - 1 more or less conventional lecture (next week) on topology - following lecture as introduction to the material sufficient for understanding the standard ruler project and together discussing how we could divide the work (code writing, debugging, paper writing) ** "Archeops - topology project?" ** Micha� asked about the possibility of writing software for topology work. This is certainly possible, but it's less clear that it would result in publications or observational detections. I think that the Grenoble people would be happy to have someone contribute to programming work for Archeops analysis, and that in return, we could do topology analysis, but this would be a *much* bigger scale project (in terms of programming effort, coordination, and... politics) than the standard ruler project. Comparison local parameter vs topology projects: definite exciting result if result, then how important standard ruler YES moderate to big Archeops topology NO big to huge effort (programming, coordination, politics) standard ruler easy in a few months, IMHO Archeops topology heavy, over 12 months Hmmm.... There is also a very easy, quick topology project which should lead to a publication relating to a principle for detecting topology. :-) Let's call this: ** "roots of the identity project" ** The ultimate decision should be from the students - I personally will do the standard ruler project sooner or later. Sooner would be better, but later is also OK. I will also do the roots of the identity project, when I have a spare moment... Think about it, discuss it on this list, and at next Friday's meeting after going through a topology introduction we can see what people think. Pozdrawiam Boud

1 0

Fri 01/03/2002 14h-16h Radio Obs Piwnice
by Boud Roukema 28 Feb '02

28 Feb '02

Reminder: shape-univ lecture tomorrow Fri 01/03/2002 14h-16h Radio Obs Piwnice Cze�� Boud

1 0

shape-univ monograph course 14:00 Fri 1 Mar 2002
by Boud Roukema 22 Feb '02

22 Feb '02

Dear students, Nazywam si� Boud Roukema, jestem kosmologiem, zacz� jak adiunkt w Centrum Astronomii. This semester, I'm giving a "monograph course" on "the Shape of the Universe", or more precisely, on observational ways of trying to measure the shape of the Universe, aimed at the level of 4th year students, but 3rd years or other students are also welcome. You'll find more information on this at: http://www.astro.uni.torun.pl/sympa/shape-univ/index.html Sorry for the confusion about the starting date! Because of this, the next session will have a subject which is essentially the same as the first, unless you all think it is too easy and you already know it! So here are the details: ---------------------------------------------------------------------- Time/Date: 14:00-16:00 Friday 1 March 2002 Place: Radio Observatory main lecture room, Piwnice Goal of first lecture: * Basic concepts of curvature, topology, extra dimension as a psychological tool to help imagine curved and/or multiply connected spaces, comoving coordinates, with the aim of making sure the 4th year students (or others relatively new to cosmology) understand these. ---------------------------------------------------------------------- Two students who were there, Pawel and Andrzej, said that it was very easy, so if they can explain everything to you, we could maybe move immediately on to observational techniques. But this depends on you. Apparently there was another course on something about business and computing in the other building this afternoon. Well, the best way to learn computing if you want to be able to find a job in it and/or want to create your own software company is in the free software movement, e.g. GNU/Linux and the GNU Public Licence (GPL): http://www.gnu.org/home.pl.html http://groups.google.com/groups?group=pl.comp.os.linux If you write software for a company which is *not* protected under the GPL, and if the company fires you and you want the freedom to use this software in another company, you will not be allowed to! Not only will you have lost your job, but all your intellectual effort will be wasted. :-( And starting your own company would be much easier with free software, since you would only have to add one small new idea to the huge amount of existing free software. You would not have the cost of buying proprietary software or using an unstable, inefficient operating system full of bugs. So free software is better, z moim skromnem zdaniem... In astronomy, most of the software we use and write is more or less free software, but its commercial value is about... zero z�oty. However, if you help work on cosmology GPL software, it would be a good learning exercise. And I think that one of the best ways for you to learn about observational measurements of the shape of the Universe is to work with software that is related to observational work on the shape of the Universe. I would be happy to support students interested in doing this, and hope that both the face-to-face lectures and the shape-univ mailing list (and archive) can be used to think about this. Anyway, the content of the course is more or less: ---------------------------------------------------------------------- = Quantitative = - curvature measurements, e.g. http://de.arXiv.org/abs/astro-ph/0201092 - topology measurements, e.g. http://de.arXiv.org/abs/astro-ph/0010185 [This is the ref I promised during today's lecture. For a *lot* more detail on topology (but slightly out-of-date regarding observational research and with a few small, minor errors) is LaLu95: http://de.arXiv.org/abs/gr-qc/9605010 ] Please look for good introductions to these on the web (w po polsku lub w po angielsku), i prosz� wysylacie any links that you think are good to the shape-univ mailing list. That way, everybody else can just click and read. = Qualitative = - curvature Here's one which seems nice, but is essentially without algebra: http://www.astro.princeton.edu/~clark/ToC.html It's said to be at high school level, but it's important to develop intuition. Understanding is not just abstract symbols on paper, it's also intuitive understanding. But please provide a better link if you can! - topology - intuition Here's a good exercise for developing topology intuition: http://humber.northnet.org/weeks/TorusGames/ Don't be embarrassed to play the games - it's part of learning. ---------------------------------------------------------------------- To subscribe to the list, send an email to sympa(a)astro.uni.torun.pl and put subscribe shape-univ in the Subject: or main body. To send an email to the list, just send it to shape-univ(a)astro.uni.torun.pl . Hope to see you next Friday, Pozdrawiam Boud

1 0

shape-univ up and running!
by Boud Roukema 21 Feb '02

21 Feb '02

<cross-post to cosmo-spotka & shape-univ> Here is the description of the first shape-univ announcement (this way this will be archived on the shape-univ archive): http://www.astro.uni.torun.pl/sympa/cosmo-spotka/2002-02/msg00004.html Several people will get a double of this message. Feel free to unsubscribe from or subscribe to one list or the other (or even both) depending on your interests, and encourage others to subscribe to those lists they feel are relevant. Hopefully see you all tomorrow! Cze�� Boud

1 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Shape-univ