I like reading emails in e.mail client as opposed to on web browsers. This have many benefits such as you can see mails from all of your account in one place, get notified as soon as there is an incoming mails and the best of all is you don’t have to type password every time you want to log in. Again, you can move mails from account to another account very easily. The biggest advantage in my view is the ability of checking mails very quickly irrespective of internet connection speed. Whenever a mail is delivered to mail-server, the e.mail client download the mail to client computer using IMAP or POP3 protocol. Client machines cache these new mails. So, the user don’t get delay of reading mails because of slow download speed. However every pros with some cons. If you don’t have a dedicated personal computer and wants to check emails on different computer then you should be better of with web browser.
Working with latex is sometimes very painful because of several packages to be included. I am building this page to use as a reference when working with pdflatex, TeXShop, or other PDF-oriented compiler. Although I am providing required package with the code, we must include all packages at the header of the tex file i.e. just after \documentclass command.
Prof. Felisa Vazquazabad showed me a cool feature in MAC. In her computer she can toggle between latex source code and the pdf file very easily. And when toggling, the particular portion in pdf is highlighted corresponding to the position where cursor is placed in latex source file. I searched for this feature and found out that this is called forward search (mapping a point in latex source to corresponding region in pdf) and inverse search. This is highly possible in Emacs-Auctex with evince that is the default pdf viewer in Ubuntu. I am using Emacs 23 with auctex 11.87 and Evince 3.4 in Ubuntu 12.04-64bit version.
Today I found another feather on the hat of Emacs. Any man who is tired of writing math formulas, tables or figures in latex would love this feature. In normal latex environment users have to write the latex source in some editor and the compile it and see the output in another file. So, all the time you make a change, you have to swap window which increases time consumption and error in writing.
I love emacs because it allows you to customize it according to your need. I am using emacs as my latex editor and found it very easy. I have talked about it earlier. Recently I found out that generating a dvi file is not a good idea from tex file. because when we generate dvi file, figures and pictures are not included in the dvi file. Instead a blank spot is provided in dvi file and during displaying it, the dvi viewer link to that particular file to that blank spot. So, I can’t mail only dvi file to someone who don’t have access to those pictures. pdf files are extremely reliable in this case. Once a pdf file is generated, it contains all information and pictures inside it. So, we don’t need any extra file alongside. Generally pdflatex is used to generate pdf file out of latex source code. It is very generous. However emacs use latex command by default which generate dvi file. If we want to run pdflatex instead of latex, we have to set some environment for emacs. (setq TeX-PDF-mode t) will tell emacs to run pdflatex when we press C-c-C-c-RET (Ctrl+c, Ctrl+c, Enter).
For last few weeks I was searching a good tool where I can plot graph / chart generated from simulated results. I found google spreadsheet really useful in that case. The problem with google graphs is that it does not look good in black and white. Again spreadsheet tools does not support single graph to be exported as picture. I tried to run gnuplt and believe me, running gnuplot successfully is a big headache. So, I switch back to plotting graph on spreadsheet such as ‘calc’ in Ubantu. Now, how can I export it? If I take screenshot and then include it in my paper, the image conversation screw up the appearance of the graph. Then I figured out that I can plot the graph in Calc, change appearance by selecting each data series and placing custom point heads, all in black and white mode, changing line types so that when printed in B/W, the chart is recognizable. Then just copy the graph and pasted it in another program called draw. Then go to export and select ‘eps’ format. In the below check the box titled ‘Selection’. This will export only the chart instead of whole page. Save it and open in Latex. Bingo… my problem is solved and I get the best image manipulation as eps formatted images are vector image and does not blur when you are zooming in.
Often we need to rename a group of files into specified format. Think of of the scenario when you download few dozones of pictures from your camera and all of those pictures named as Image00000123.jpg. Would not it be a good idea to rename all those pictures starting with specific name such as NiagaraTrip001.jpg and so on? Well, you can do it by renaming each picture manually but who wants to make his hand dirty when it can be done in a very sophisticated way by writing one line of command in Linux shell?
Latex program reads from input text (.tex) file and several other (e.g. jpg, eps etc.) files to produce a .dvi file. .dvi files encode information on fonts and formatting and are machine independent. DVI files can be then converted to many formats like pdf, postcript, etc. .tex files are created as “What You See Is What You Get.” Text formats are not displayed in input file but rather formatted display would be generated in the output .dvi file.
Nowadays, I am writing some code to simulate Erald Troja’s research on a wireless network. We can use standard simulators but the difficulty in that is to plug new algorithms into an already developed protocol suit. So, we decided to use any programming language and simulate the network treating it as a graph, i.e. vertex as mobile nodes and edges as wireless links. Every edge has many time varying characteristics depending on its location.
If you are in the phase of starting your research in Network or Wireless domain then, choosing correct simulator can save a lot of your effort. I am writing this blog according to my own opinion and it should not be taken as something standard or reference. A list of well known simulators are provided here. Some of the simulators are developed and maintained by academia / research community while some are maintained by commercial enterprises.
Nowadays, modern operating systems such as Windows 7 and Ubuntu come pre-installed with Bangla fonts. Do you remember old systems where reading an article written in Bengali Unicode was a big headache (most-likely we see stream of )? I am not a person involved with text processing, but here I am expressing my view only.
When I think about my past days of doing silly programs I sometimes laugh on me, how stupid was I, using one after another editor for writing programs. Sometimes I was searching for IDEs exhaustively. I remember, in 2005 started programming C on Turbo C++ IDE (the famous blue screen idiot) and most of the time I used to play one audio CD (composed by Tabun containing some old RD Burman songs). Then I used Eclipse for Java which was a very smart IDE in deed. Then while teaching at IEI, I had to use a very lightweight compiler suitable for my netbook and I opt Genie this time. In the first year of my PHD I learned Haskel and Prolog for which I used their custom interface.