Entries from Stephan Paukner

Thursday, December 7. 2006

Tätigkeitsbereiche

Um einmal niederzuschreiben, was ich bei der SOL4-IT alles so gemacht habe, habe ich mir eine Liste meiner Tätigkeiten erstellt. Das wäre ein Grundgerüst für die nächste Bewerbung. Um nicht zu viel zu verraten, habe ich hier die jeweiligen Kundennamen nur abgekürzt.

Beschäftigungszeitraum: 10.07.2003 - 31.01.2007.
Alle Aufgabenbereiche wurden in direktem Kontakt mit dem jew. Kunden abgewickelt.

Lotus-Notes-Development

Ab 2004/06: S. S. AG: Einführung eines DMS unter Analyse und Abbildung der Kundenworkflows. Umsetzung umfangreicher Anpassungen und Erweiterungen. Implementation und Integration einer Artikel- und einer Projektverwaltung. Erweiterung des Helpdesksystems. Projektaufwandsanteil >110MT.
Ab 2005/04: Implementation und ständige Weiterentwicklung des hauseigenen Produkts “Visual Team Calendar” in Kooperation mit der AdHoc, s.r.o. (Prag, CZ). Anpassungen für C. Z. GmbH (Geräteverleih), A. A. und A. A. GmbH.
2004, 2006: E. Reiseversicherungen: Implementation einer automatisierten Verarbeitung von Attachments eingehender Mails, Sortierung nach konfigurierbaren Kriterien, Datenextraktion. Erweiterung um FTP-Transfer.
A. A. GmbH: Übernahme der Betreuung des DMS. Migration des angepassten DMS auf neue Version. Implementation und Integration einer Verkaufsunterlagen-Verwaltung.

Linux-Systemadministration

Teilhabe an der Hauptverantwortung für den Betrieb des Webhostings, verteilt auf 4 IBM-Rechner: DNS von über 150 Domains, ca. 140 Webauftritte, ca. 330 User, Mailserver mit Virus- und SPAM-Filter, J2EE-basierendes CMS, Lotus-Domino-Hosting.
Installation und Erweiterung eines Überwachungssystems inkl. differenzierter Alarmierung bei Ausfällen.
Migration (Konsolidierung) von PostgreSQL auf MySQL.
Wartung der firmeninternen Systeme.

Kundenprojekte:

A. A. GmbH: Selbstständige Vorbereitung und Abhaltung einer fünftägigen Linux-Schulung für 4 Personen.
F. H. GmbH: Wartung und Erweiterung des Mailservers durch SPAM-Filter.
R. F. GmbH: Systemmigration auf RAID-Festplattenverbund, Backup-Implementation mit Transfer auf IBM AS/400-Rechner.
Dr. K. & M. GmbH: Systemmigration auf RAID-Festplattenverbund, Lotus Domino Backup-Szenario.
S. S. AG: Migration eines J2EE-basierenden CMS inkl. Webauftritt von Suse Linux auf Debian GNU/Linux.
M. Handelsges.m.b.H., G. & H. OEG: Migration eines Windows-Fileservers auf Linux/Samba.

Web-Development

2003: Wartung, Betreuung und Erweiterung des Immobilienportals FMH (G. & H. OEG) in PHP/MySQL, z.B. Erweiterung der Suchfunktion.
2004/03-2005/03: Wartung, Betreuung und Erweiterung des Hotelportals H4Y (E. Verlagsges.m.b.H.) in PHP/MySQL und tw. JavaScript/DOM, z.B. Mehrsprachigkeit, Mitgliederbereich, Newsletterkampagnen, Veränderung der Reihung und Hervorhebung von Suchergebnissen, Erweiterung des CMS, Bannerverwaltung, Bannerrotation.

Posted by Stephan Paukner in Information Technology at 12:07 | Comments (0) | Trackbacks (0)

Defined tags for this entry: german, work

Tuesday, December 5. 2006

Gleam of hope

I stop working at SOL4-IT at the end of January 2007. From that time on I can work 100% on my Master thesis! But I don’t have to live from my savings: I’ll apply for a final degree scholarship (Studienabschluss-Stipendium). For 6 months I will (or could) be promoted with €1000 every month, with an additional (unpaid) timeframe of 6 months until I’d have to pay the amount back if I hadn’t finished by that time. I’ll not be allowed to hold any job during that phase. That gives me everything I need for a scientific work.

Posted by Stephan Paukner in Master's Thesis at 08:56 | Comments (0) | Trackbacks (0)

Sunday, November 26. 2006

jpg document vs. JPEG image

Linux sucks on the desktop. I’ll write a rant on this soon.
In Nautilus, suddenly, one day, after a general (apt-get) upgrade, there were no more thumbnails generated for new JPEG images. When I clicked one of the icons, instead of opening the image in the viewer ‘eog’, a message raised, saying

The filename “IMG_1234.JPG” indicates that this file is of type “jpg document”. The contents of the file indicate that the file is of type “JPEG image”. If you open this file, the file might present a security risk to your system.

followed by the usual security-blah-blah. I couldn’t examine what the cause was, until I finally stepped over a posting telling that the file ~/.local/share/mime/globs contains an overfluid entry. Remove that entry containing the string ‘jpg’.

Posted by Stephan Paukner in GNU/Linux at 19:31 | Comments (2) | Trackbacks (0)

Saturday, November 18. 2006

Convolutions

(Q26) Meanwhile, I know that a convolution of one function $f$ with another function $g$ means that $f$ will be “ $g$ -smeared”, and vice versa. I visualized some convolutions using Octave. I was prepared to implement $f*g(k)=\sum_jf(j)g(k-j)$ for two vectors, but then stepped over HGFei’s conv2fei which makes the surprising use of the identity $\widehat{f*g}=\hat f\cdot\hat g$ ; I wouldn’t have come to the idea to do it this way, you only have to multiply two FFT’s and then do the inverse FFT.

First I convolved a square with a small Gaussian—this is nothing less than a Gaussian blur:

$*$

$=$

Note that the blur would already occur even if the second object were solid! Look at the square convolved with itself:

$*$

$=$

Then I watched how these bars convolve with each other:

	$*$		$=$
	$*$		$=$

Notice how the blurness shrinks when the two objects can’t overlap much.

Posted by Stephan Paukner in Master's Thesis at 16:48 | Comments (0) | Trackback (1)

Defined tags for this entry: mathematics

Raising questions, 66-69

What are Fourier multipliers?
What are Frechet spaces? Why are $\mathscr S$ and $\mathscr S^\prime$ not Banach, but only Frechet?
Is there a difference between $C_0$ and $C^0$ in HGFei’s notation?
Why is $\langle f,g\rangle Th=\langle f,Tg\rangle h$ ? (There’s a relation to Q2.)

Posted by Stephan Paukner in Master's Thesis at 16:29 | Comments (0) | Trackbacks (0)

Defined tags for this entry: mathematics

Wednesday, November 15. 2006

The exponential in $R^d$, II

(Q15) I’m gonna show you some cool Gabor atoms on $\bbR^2$ . First, the function $u_\omega(x):=e^{\pi\i\langle x,\omega\rangle}$ , as described previously, indeed looks like this:

CODE:

[xx yy] = meshgrid(linspace(-2,2,100)); v=[1 1.5]; e1 = exp(pi*i*( v(1)*xx + v(2)*yy )); imagesc(real(e1)) w=[-1 4]; e2 = exp(pi*i*( w(1)*xx + w(2)*yy )); imagesc(real(e2))

See how the value of the frequency changes with the length of $\omega$ , and the direction with the orientation of $\omega$ , just as described previously.
In the STFT, these frequencies get reduced locally by an “envelope function”. One could take the Gaussian window $e^{-a^2x^2}$ to achieve this:

CODE:

g1 = exp(-(xx.^2+yy.^2)); imagesc(g1) g2 = exp(-4*(xx.^2+yy.^2)); imagesc(g2)

And now these are the modulated Gaussians, whose set of translates across $\bbR^2$ forms the building blocks for Gabor analysis on $\bbR^2$ :

CODE:

imagesc(real(g1.*e1)) imagesc(real(g2.*e2))

Posted by Stephan Paukner in Master's Thesis at 17:15 | Comments (0) | Trackback (1)

Defined tags for this entry: mathematics, matlab, octave

Tuesday, November 14. 2006

Raising questions, 63-65

What is $\Lsp^2_\mathrm{loc}$ ?
What are singular matrices?
What is the condition number of a matrix?

(Continued from questions 1-62)

Posted by Stephan Paukner in Master's Thesis at 20:39 | Comments (0) | Trackbacks (0)

Defined tags for this entry: mathematics

Sunday, November 12. 2006

The exponential in $R^d$

In question 15 I want to make a picture how $u(x,y):=\langle x,y\rangle$ looks like for $x,y\in\bbR^d$ . I want to take at least two-dimensional vectors, as the inner product is too trivial otherwise. But then $u$ is a function on $\bbR^4$ , which is rather difficult to plot. So I want to fix $x=a$ and plot $y\mapsto u_a(y):=\langle a,y\rangle$ , which is then a function on $\bbR^2$ with values in $\bbR$ and therefore easy to plot. As the inner product is $\sum_k x_k y_k$ , it is in the 2-dimensional case a simple plane-equation $z=a_1y_1+a_2y_2$ ; of course, the inner product is linear and goes through the origin! It cuts the zero-plane $z=0$ in an angle which is dependent on the orientation of the fixed $x=a$ : The zero-line evolves where the vectors $y$ are orthogonal to the fixed $x$ , i.e., $\langle x,y\rangle=0\Leftrightarrow x\perp y$ .

This zero-line will now change its angle on the zero-plane while varying the fixed vector $x$ . This gives a picture how it behaves as a whole. In Octave, I set

[xx yy]=meshgrid(linspace(-2,2,50));

and then made a plot corresponding to $x=\left(\begin{smallmatrix}1\\1\end{smallmatrix}\right)$ by

mesh(xx+yy)

The resulting 3D-plot shows a boring inclined plane. To better see the location and orientation of the zero-line, I plot the absolute values:

mesh(abs(xx+yy))

To see the orientation even better, I chose

imagesc(abs(xx+yy))
imagesc(abs(xx+2*yy))
imagesc(abs(-xx+4*yy))
corresponding to the further vectors $\left(\begin{smallmatrix}1\\2\end{smallmatrix}\right)$ and $\left(\begin{smallmatrix}-1\\4\end{smallmatrix}\right)$ . This is what I get:

The turning around of that plane now corresponds to the behavior of the mentioned inner product function on $\bbR^{2d}$ . (Beware that the absolute values used in the plots hide the asymmetry of the inner product function!) The values are always real and go into the exponent of the exponential function. $e^{\i\langle x,y\rangle}$ then “rotates fastest” as $y$ runs along to a fixed $x$ , and is constant = 1 if it runs orthogonally; Beware that there’s a difference whether it is constant = 1 or if it reaches that value again and again. So, in the integral of the FT, the exponential function simply covers all frequencies via exhausting all values of $x$ , and the FT is then a function in $\omega$ , which takes out a single orientation of those frequencies; As already mentioned some time ago, the term “frequency” isn’t useful anymore in higher dimensions. For two-dimensional signals, i.e. images, a “frequency” can be interpreted as a pattern of lines: The narrower the pattern, the higher the frequency is. The difference to one-dimensional signals is, that those frequencies may now have an angle. It is not enough to just look at a single frequency of that line-pattern, but also to all their orientations! Only with using all orientations of patterns, an image can be analyzed according to the contained frequencies. The 2D-FT of an image therefore tells you what patterns occur in what orientation. To be more precise, you select the frequency by the length $|\omega|$ and the orientation by its angle $\arg\omega$ , because the inner product as a function in $x$ grows faster with growing length of $\omega$ . For me, this is now a milestone in understanding the 2D-FT.