Re: [Ilugc] How to run a shell script in windows

[Kapil Hari Paranjape]

Hello,

On Fri, 04 Dec 2009, Balachandar wrote:
>I know that the shell script can automate many works in linux 
> environment.If i 
> want the same in windows how can i do? Is there anyway to run shell scripts 
> in 
> windows environment ?

There is are ports of the bash/zsh/pdksh shells to the cygwin
environment of windows. Note that you will also need the ports of the
standard utilities which make the shell so powerfull under Unix.
These are also mostly available.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Memory Test In Ubuntu 9.10

[Kapil Hari Paranjape]

Hello,

On Wed, 02 Dec 2009, Hrishikesh Murali wrote:
> On 12/2/09, Kapil Hari Paranjape  wrote:
> > Disable graphics mode while running memtest.

> I switched to console mode and then tried running memtest86+. It's working :)
> I guess its a bug in the graphical mode of grub2. Thank you anyway :)

The point is that memtest assumes that it is running in text
mode.

Perhaps grub2 should provide a shift back to text mode for running
programs that do not support a framebuffer.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Memory Test In Ubuntu 9.10

[Kapil Hari Paranjape]

Hello,

On Wed, 02 Dec 2009, Hrishikesh Murali wrote:
> Now, when I press  after the computer hangs, the computer
> reboots. Why I am saying this is because  is used for rebooting
> the system when the memory test is going on. So, is it that the memory
> test is running, but I am not able to see anything on the screen? Is
> there a workaround for this bug?

Disable graphics mode while running memtest.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (dd - split a file into multiple pieces with bad internet uploads)

[Kapil Hari Paranjape]

Hello,

On Mon, 30 Nov 2009, Girish Venkatachalam wrote:
> Long ago I sent a patch to add resume feature to scp(1)

One alternative to these suggestions is to use 'rsync' with the
'--partial' flag.

Who knows, if two blocks of your file are the same then you may even
save on bandwidth compared with 'scp'!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] cp command

[Kapil Hari Paranjape]

Hello,

On Thu, 26 Nov 2009, Sagayaraj wrote:
>   How to copy the specific files only in cp command in RHEL 5 or all linux
> flavor?
> Ex i've 1000 files in a folder, i want to copy only 715 files olny in that
> folder to paste other place. what command using for that..waiting for your
> prompt reply.

mv $(ls | head -715) new_folder/

--
___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] How to generate a GPG fingerprint for my name ?

[Kapil Hari Paranjape]

Hello,

On Wed, 25 Nov 2009, Balachandar wrote:
>I noticed that many peoples are listed their GPG Fingerprint for their 
> name in mails.I want to know how to create one for my name.I googled about 
> it.But couldnt find information about this.Thank you.

The GPG fingerprint is not for the name but for the public key.

Read the online documentation about GNUPG for more information.
 http://www.gnupg.org/
(There is a mini-Howto under documentation/HOWTOs).

Keep your private key safe!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (script - record input and output from

[Kapil Hari Paranjape]

Hello,

On Mon, 23 Nov 2009, Suresh Kumar wrote:
> How about logging the ssh operations. will  script command works?
> I mean,
> 1) Login to another linux box using ssh
> 2) doing some operations (ran some commands)
> 
> How do i log this using linux command  (script)

Did you try the obvious method?
 $ script /var/tmp/remote_ssh_session
 $ ssh remote_host
  login with some commands to remote session and logout
 $ exit

Don't forget to remove your password/passphrase from the script
file if you are not using out-of-band authentication[1] for ssh.

Regards,

Kapil.

[1] Out-of-band authentication for ssh means that ssh uses a
public-key which is unlocked in a different terminal session from the
one used to make the connection. This is managed with ssh-agent or
gpg-agent etc.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] IBM and the indian debacle

[Kapil Hari Paranjape]

Hello,

On Sun, 22 Nov 2009, anirudh singh shekhawat wrote:
> IBM should not support open source if they cant simply stick to it
  ^^
> in a simple student's competition.

Really? You want IBM to _stop_ supporting open source just because
some "suit" somewhere in this organisation has a short term viewpoint?

Note that most large companies like IBM and Tata are bureaucracies in
the same way that large governments (like GoI) are.  So it is quite
possible for the left hand to oppose the actions of the right hand.

> And what more can we expect from colleges authorities if big
> organizations like IBM are slapping open source right there in front
> of everyone,

For one, I _expect_ someone who is called a "teacher" to have a more
independent viewpoint than someone who works for a company.

I understand your annoyance at the fact that some students will get
sucked into creating proprietary code. Moreover, some other students
who are more "into" open source will be left out in the cold. To them
I would say, reserve your energy for programmes like GSoC which do
seem to mostly do the right thing.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] command completion

[Kapil Hari Paranjape]

Hello,

On Sat, 21 Nov 2009, steve wrote:
> (*) for the curious and the cautious -- bash-completion complements
> the regular  completion feature of bash by 'intelligently'
> completing arguments, for eg: only directories would show up after a
> 'cd' and only pdf files listed with a 'evince'
 ^^

I think what you meant is:

 only pdf,djvu,dvi and other graphic files listed with a 'evince'

'evince' is a pretty capable viewer!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] command completion

[Kapil Hari Paranjape]

Hello,

On Fri, 20 Nov 2009, Vamsee Kanakala wrote:
> If you're using bash, hit ctrl+r and start typing the first few
> chars, hit enter once to copy the command, again to execute.

Also look at what 'Esc+Tab' does for you. Sometimes that is what is
wanted --- to re-use an earlier command with different options.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Re: Fedora 12 Lets Users Install Signed Packages, Sans Root Privileges

[Kapil Hari Paranjape]

Hello,

On Thu, 19 Nov 2009, Mohan R wrote:
> Salvadesswaran P.S. wrote:
> >Now I can tell my
> >friends that they can install the packages they need for any purpose.
> >They are prone to being too lazy to enter the root password to install
> >package(s), like most normal desktop users.
> 
> Don't ask these kind of lazy friends to get into Unix for the
> namesake of ease of use. Unix is different, please let it be what it
> is.

I thought of writing the same thing and then realised --- perhaps
these lazy people will become less lazy once they enter the dynamic
world of *nix. Some lazy people start taking evening walks once they
live in front of a park!

> I was a Ubuntu user and recently switched to Debian,

Welcome!

> This is not the first time fedora is in controversy regarding
> security. Recently fedora's repositories were compromised because of
> bad signatures in its packages, check the below link.

If your remember events from a few years ago, there was a major
problem in the international SSH usage system because of Debian. So
we cannot claim to be totally pure either!

Not only is there "more than one way to do it" but we can learn a lot
by seeing how the other team does it. If we flame other distros to a
crisp we will not have the opportunity to learn from their mistakes
instead of ours!  :-)

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] gmail backup on debian

[Kapil Hari Paranjape]

Hello,

On Wed, 18 Nov 2009, Raja Subramanian wrote:
> I want to back up my gmail account mails to my local
> debian box.  Preferably something that can also save
> gmail labels.  I can setup squirrel mail to read-only
> access the locally available email.

I use "offlineimap" for accessing my gmail account and other IMAP
accounts. It is quite easy to setup.

However, there are some problems that should be mentioned.

The imap.gmail.com server converts each label into a new imap folder.
Thus, each such folder is downloaded by offlineimap causing a mail
with multiple labels to be downloaded many times. I don't know of any
software that can work around this.

If you want to automate the process then the configuration file for
offlineimap must contain your gmail password which is not terribly
secure. It would be nice if there was an interface that could use a
keyring or gpg-agent or some other dynamic password store.

The local store is maildir but I think this is a feature and not a
problem!

The nice thing is that the program is written in python so it may be
possible to fix these things with some scripting!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Brazen OpenBSD advocacy

[Kapil Hari Paranjape]

Hello,

On Tue, 17 Nov 2009, Girish Venkatachalam wrote:
> Interestingly one has Rajni and the other Gauthami. ;)

You realise, of course, that you may be under risk from copyright
lawyers who might ask how you obtained these copyrighted images!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (cygwin - POSIX environment on Windows)

[Kapil Hari Paranjape]

Hello,

On Tue, 17 Nov 2009, Senthil Anand wrote:
> On Tue, 2009-11-17 at 08:43 +0530, Girish Venkatachalam wrote:
> > You can install Cygwin from http://www.cygwin.com
> > 
> > Whenever I am forced to use Windows I do this.
> > 
> 
> Another option would be Interix which now called Subsystem for
> Unix-based Applications (SUA) from Microsoft. It is a 32/64 bit UNIX and
> POSIX subsystem for the NT kernel similar to the Win32 subsystem. 
> 
> When I worked with it a few years ago it was based on the OpenBSD
> userland. Its main advantage was that it was a native subsystem unlike
> Cygwin which was POSIX on top of Win32. Its main disadvantage is that
> the latest version (6, 6.1) needs Vista/Windows 7 Ultimate or
> Enterprise.

Its other important disadvantage of SUA is that you do not have access
to the source. So if it has a bug you can't easily find a fix for it!

Secondly, you probably cannot distribute GPL binaries linked against
it since that would most likely violate MS licenses.

Cygwin or other such systems have a distinct advantage there.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (cksum utility in OpenBSD)

[Kapil Hari Paranjape]

Hello,

Sorry for the duplicates. The connection here is very poor and though
the mail was sent, I received an error message so I sent it again!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (cksum utility in OpenBSD)

[Kapil Hari Paranjape]

Hello,

On Mon, 16 Nov 2009, Girish Venkatachalam wrote:
> Basically RSA relies on the NP complete problem of prime number
> factorization and Diffie Hellman relies on discrete log problem.

First of all, I should clarify that my aim is not to try to prove
anyone wrong. However, when a statement is made on a public channel
that is used by newbies and this statement may (IMNSHO!) mislead some
of them, I feel that it should be corrected.

The case in point: factorisation is not known to be an NP complete
problem.

However, people who study algorithms do believe that it is
unlikely that there is a P algorithm for factorisation.

A P algorithm for factorisation is one which will factorise a large
number N in log(N)^k steps for some fixed k (independent of N).

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (cksum utility in OpenBSD)

[Kapil Hari Paranjape]

Hello,

On Mon, 16 Nov 2009, Mano wrote:
> On matters mathematics I would trust Dr. Kapil more than anyone else
> on this list!

Never (at least almost never) _trust_ someone else's mathematics. The
whole point of mathematics is that if it is correct you should  be
able to verify things for yourself. A bit like open source --- if the
program has a flaw you can check it!

In this case, perhaps I interpreted incorrectly what Girish had to
say. My only justification is his obtuse style of writing! :-)

Though this is a little OT now, I am explaining this since Girish has
no time!

Here is a brief description of RSA which ignores some technical
points:

 Take a number N which is a product of two (large primes) p and q.
 Take a smallish prime r like 23 or 31 which does not divide (p-1) or
 (q-1). Using p, q, r you can calculate s so that (p-1)(q-1)
 divides (rs-1).

 Your public key is (N,r). Your private key is (N,s).
 It is assumed that the p and q are not known to anyone except the
 person who holds the private key.

 Encryption takes m | 0 < m < N and takes it to (m^r modulo N).

 Decryption takes n | 0 < n < N and takes it to (n^s modulo N).

 One can show that (m^(rs)-m) is divisible by N providing that m is
 different from p and q. 

I hope this explains my remark that encryption and decryption are
inverses of each other.

Here is a brief description of Diffie-Hellman key exchange which also
ignores some technical points:

 Let P be a large prime and g | 1 "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (cksum utility in OpenBSD)

[Kapil Hari Paranjape]

Hello,

On Mon, 16 Nov 2009, Girish Venkatachalam wrote:
> In fact the RSA algorithm is based on this message digest algorithm
> but I don't think many people recognize this fact.

Which RSA algorithm?

RSA encryption does not require a message digest algorithm.

The RSA signature algorithm usually uses message digest algorithms for
efficiency and for a few other reasons.

However, the RSA encryption and decryption algorithms are inverses of
each other so one could make a signature by applying RSA decryption to
a message text. Only the person having the private key is capable of
doing this!

The recipient would then apply RSA encryption (since the public key is
public!) to read the message and if this works then it verifies the
sender as well!

> RSA is basically an implementation of this concept a^b^c = a ^c^b =
> a^bc with modulo N thrown in.

I think you are confusing the Diffie-Hellman key exchange with RSA.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] archiving/sharing mail for future reference

[Kapil Hari Paranjape]

Hello,

On Sun, 15 Nov 2009, Satheesh Kumar wrote:
> "We need to archive certain individual mails so that it shall be
> available for future reference for any within the organization. i.e
> the user should be able to send the mail (along with the attachments),
> this mail has to be published as web page/wiki so that it is available
> to anyone in the organization for future reference".

The requirement is a bit vague so a number of a solutions are
possible. For example, using mhonarc as follows:

 1. A specific e-mail address is reserved for this system.
 2. Only designated people (i.e. "From:" addresses) are allowed to
 send mail to this address.
 3. The mailbox is processed periodically using cron.
 4. The periodic process runs mhonarc to create the web page(s).

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Linux for older people

[Kapil Hari Paranjape]

Hello,

This is probably getting OT. However, ...

On Sat, 14 Nov 2009, Kenneth Gonsalves wrote:
> nothing to beat MSI WIND netbook which comes with opensuse - everything works 
> out of the box, and a huge number of applications are there. costs 17K.

The Acer is _lighter_ which may may a difference for older people.

One other interesting machine on offer is the Samsung NC 110 (or is it
120 or 138?) which offers 9-11 hours of battery life. This means that
you need not lug around the charger during the day.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] ed cheat sheet

[Kapil Hari Paranjape]

Hello,

On Sat, 14 Nov 2009, Girish Venkatachalam wrote:
> I shall run a tip on ed soon but it is not for the faint hearted.

Have a look at 'edbrowse' while you are at it. It also allows you to
access the internet through the 'ed' editor interface.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Carte Blanche - Computer Society of MIT

[Kapil Hari Paranjape]

Hello,

On Tue, 10 Nov 2009, sivakumar bharadhwaj wrote:
> suggestion : for wooing new people for FOSS.

Good to hear that you are thinking about this!

> 2. To show them how to install a linux distro in a PC (including dual
> booting) - step by step (if possible in printed format)

 "Install once, run many times!"

Hence it is not clear to me that showing novices the steps of
installation is really helpful.

> 4. show them how to use yahoo messenger (gYachi) and Skype - in FOSS
> software.

Skype is not free software. However, showing that proprietary
software can also be run when needed by the person concerned is
certainly helpful.

> 8. Printed sheet - having a lot of contacts, numbers, names, - who they can
> call for assistance (we can later make them as volunteers who will help
> other people, after using FOSS).

Also put in web sites and other sources of information!

> 9. keep around 20 to 30 OLD PCs with FOSS software - for anyone to test, and
> use internet etc.

This is a nice idea with one small problem --- space/location.
Where will you keep these demo machines?

> hope you all dont laugh at these.

Not at all. Keep all the ideas flowing!

Kapil.
--
___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] qt in debian sid

[Kapil Hari Paranjape]

Hello,

On Sun, 08 Nov 2009, Abishek Goda wrote:
> I am using debian sid and I tried upgrading qt for a python-qt based
> app. since the upgrade qt has been segfaulting. one mistake i did is
> not noting what versions of qt were there before the upgrade.

Have you checked with bugs.debian.org/qt4-x11?

By the way, the safer way to work with Debian sid is to install it
under a chroot and use it with pbuilder or schroot. (For GUI stuff
you may want to add Xephyr to the mix). This way you don't bork your
working system if sid has broken something. (In the "Toy Story", sid
was the kid who would break all the toys!)

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day (nail/mutt/mail - send mail from UNIX command line)

[Kapil Hari Paranjape]

Hello,

I almost decided not to respond to this but later thought that someone
might confuse restraint with acquiescence!

On Wed, 04 Nov 2009, Girish Venkatachalam wrote:
> For now please stick to Postfix or Sendmail. Both are bad. Postfix
> does not have a good license
>  and sendmail is too complex. Exim of course is worse.

Please distinguish opinion from facts.

As Kumar said earlier, your tips are indeed welcome but your
vehemance is merely distracting.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day (nail/mutt/mail - send mail from UNIX command line)

[Kapil Hari Paranjape]

Hello,

On Wed, 04 Nov 2009, Girish Venkatachalam wrote:
> Gilles Chehade is
> developing a SMTPD implementation but it will take around 6 more months I 
> think.

> For now please stick to Postfix or Sendmail. Both are bad. Postfix
> does not have a good license
>  and sendmail is too complex. Exim of course is worse.

> Anyway this is what you need to do to send mail assuming that your MTA
> does its job.

If all you want to do is _send_ mail by connecting to another SMTP server
then why do you need to run these smtp daemons?

Just use something like "msmtp". 

For example, to send mail via gmail using mutt I have:

 1. In .muttrc I have 'set sendmail = '/usr/bin/msmtp -a gmail'

 2. In .msmtprc I have:
 account gmail
 host smtp.gmail.com
 auth on
 user your_gmail_id_h...@gmail.com
 password enter_your_password_here
 port 587
 from your_gmail_id_h...@gmail.com
 tls on

Similar settings can work for other smtp daemons that allow remote
access.

The nice thing is that this does not need root privilege. The only
thing required is that your account can make outgoing network
connections on port 587 to smtp.gmail.com

Hope this helps you avoid things you do not seem to like!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] How to generate /dev/ttyACM port

[Kapil Hari Paranjape]

Hello,

On Wed, 04 Nov 2009, Kapil Hari Paranjape wrote:
> If you run "udev"

I should have said:
 If your system runs "udev" ...

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] How to generate /dev/ttyACM port

[Kapil Hari Paranjape]

Hello,

On Wed, 04 Nov 2009, Sudhir Gandotra wrote:
> In my system, I do not have /dev/ttyACM (0 to 4) ports.
> Tata broadband photon usb needs this port.
> How do I create/generate this port ?

If you run "udev" and your system autoloads the module "cdc-acm" then
this device node will be automatically created.

Otherwise, try to load this module and watch /var/log/syslog to see
whether any device was detected.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (ssh port forwarding)

[Kapil Hari Paranjape]

Hello,

On Fri, 30 Oct 2009, Girish Venkatachalam wrote:
> LOCAL port forwarding:
> 
> $ ssh -L 1234:127.0.0.1:4321 61.1.1.2
>  ^^^
>   
> will forward the local port 1234 to 6.1.1.2 4321.

This is "almost" true. The correct statement is that this will forward
connections made on the local machine at port 1234 to the remote
machine at the address 127.0.0.1:4321 _on_ the remote machine.

Usually, network daemons listen on *:4321 (where * takes all possible
local interface addresses) so this should work exactly as if the
connection was made to 6.1.1.2:4321.

Sometimes ... you may get a surprising result :-)

So if you really wanted 61.1.1.2:4321 then that is what you should
use.

  $ ssh -L 1234:6.1.1.2:4321 61.1.1.2

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Whitehouse.gov switch to Drupal

[Kapil Hari Paranjape]

Hello,

On Tue, 27 Oct 2009, steve wrote:
> I doubt however, that the contract for making government websites
> would ever go to private sector vendors who would likely do a better
> job,

Not in my experience! Have a look at the web sites of private telecom
companies or various banks. While their "looks" are perhaps slightly
better, their "feel" is as 90's as it gets. Full of flash and
cumbersome navigation based on un-necessary use of Javascript.

The core audience addressed by both sets of web sites is the same ---
the bulk of our ignorant upper middle class who can just about use
Internet Explorer as it came with their pre-installed system.

If a program/web-site is only designed with the generic "customer" in
mind this is what you will get --- something that _only_ caters to the
least common demoninator. Obviously programs should not be designed
with _only_ the "3l33t hax0r"s in mind either! An appropriate balance
has to be found.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (gcc -Wall -g options)

[Kapil Hari Paranjape]

Hello,

On Mon, 26 Oct 2009, Rahul Sundaram wrote:
> In RPM based distributions, the build system generally creates a
> separate debug/debuginfo package that you can install to get the debug
> symbols rather than recompile anything.

... and later ...

> Debian and derivates don't do this yet but apparently a major goal for
> the next release.

Until this goal is reached ... you can do the following.

 # apt-get build-dep pkgname
 $ apt-get source pkgname
 $ cd pkgname-version
 $ DEB_BUILD_OPTIONS=nostrip dpkg-buildpackage -uc -us -B

This will create a debug package that you can install and run
to get a stacktrace. (Warning: this debug package will overwrite your
existing package unless you make some other changes. However, you can
revert to your existing package with 'aptitude reinstall pkgname').

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (wcalc - the great grandfather of all console calculaters[math tool])

[Kapil Hari Paranjape]

Hello,

On Sun, 25 Oct 2009, Girish Venkatachalam wrote:
> wcalc  is a command-line calculator designed to accept all valid mathematical
> expressions. It supports all standard mathematical  operations,  parenthesis,
> brackets, trigonometric functions, hyperbolic trig functions, logs, and bool‐
> ean operators.

Inspired by this e-mail I installed wcalc. It _is_ a truly useful
tool for those looking for a replacement for a desktop calculator.

> It will give bc(1) or any other tool a run for the money.

However, this is false! The reason is that "bc" is a _programmable_
calculator. You can define functions in "bc". As far as I could tell,
"wcalc" is missing this.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: Subject: Re: Subject: Re: [Ilugc] [TAMTRANS]

[Kapil Hari Paranjape]

Hello,

On Sat, 24 Oct 2009, Prem Kurian Philip wrote:
> However, the reality is that those who are going to be using a
> computer will at some point or the other have to invent new words
   ^^
> for disk, internet and so on because there are no words for this in
> Tamil/any other indian language.

I thought your point is that such words should _not_ be invented and
the existing English words used instead. This is a sentiment that I
wholly support. A language often borrows word from other languages for
concepts that it is borrowing.

Later on as new concepts (about computers) are invented by people
who speak Tamil (or other Indian languages) they will use concepts
from these languages to make words for these new inventions and we
will be only too happy to see these words entering languages like
English. After all "dosa", "chutney" and "tikka" are now considered
English words.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-language-tip-a-week. test run [ LANGUAGETIP - basic overview ]

[Kapil Hari Paranjape]

Hello,

Making an assertion of intent like this:

On Fri, 23 Oct 2009, Girish Venkatachalam wrote:
> Obviously I don't want to get into an argument.

... is incompatible with *strong* assertions like this:

> All scripting languages...by that I mean *all* scripting languages
> allow dynamic typing and weak typing.
>
> This is a key point that everyone has to understand.

What do you mean by a scripting language? Is C a scripting language?
You may be aware of CSL ?

The key point to understand is that _everyone_ has their own language
biases^Wpreferences.

Since your intent is to educate us about Lua (or other languages) it
is probably most practical to:

 a. Give some examples of how useful programs are written in Lua (you
 have already done this ... so thanks!).

 b. Provided "counter"-examples of how such programs are written in
 other languages which illustrate differences from the Lua way of
 encoding the algorithm.

Let us worry about terminology and epithets afterwards.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (lua programming - Python like scripting language [ Part II ])

[Kapil Hari Paranjape]

Hello,

On Thu, 22 Oct 2009, Ashok Gautham wrote:
> On Thu, Oct 22, 2009 at 7:21 AM, Kapil Hari Paranjape  
> wrote:
> 
> > Or you could always use "socat".
> >
> > A simple "echo" over UDP could be implemented as follows:
> >  $ socat - UDP:loopback:2000,bind=:2000
> 
> This was just a simple example program to teach a language. Much like
> the "Hello World" :P (where echo "Hello, World" would always be enough)
> 
> I think the crux of this email was that lua makes it small and simple :)
> and how we write Lua scripts.

My mail was not _meant_ to detract from these merits of the original
tip-a-day. I'm sorry if that was the impression I gave.

> (It did not say anywhere that it couldn't be done otherwise)

You don't have to tell me that! I have often used 'sed' where 'tr' or
'cut' would have done the trick. TMTOWTDI as Larry Wall says.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (lua programming - Python like scripting language [ Part II ])

[Kapil Hari Paranjape]

Hello,

On Thu, 22 Oct 2009, Girish Venkatachalam wrote:
> Before that in case any of you wondered this is how you would run it.
> 
> $ lua echosrvr.lua
> 
> In another terminal
> 
> $ lua echoclnt.lua
> hi
> 
> will get printed on both sides as the text will be echoed back thro' the UDP
> socket.

Or you could always use "socat".

A simple "echo" over UDP could be implemented as follows:
 $ socat - UDP:loopback:2000,bind=:2000 

socat is a multipurpose data relay. See
/usr/share/doc/socat/EXAMPLE.gz for more.

By the *nix philosophy socat should have been written at the same
time as named pipes and sockets were created.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

[Ilugc] Handling mailing lists - part 1

[Kapil Hari Paranjape]

Hello,

Since handling mailing lists was referred to in my earlier mail here
is "part 1" of a tutorial on handling mailing list traffic. I hope
other parts will be written by others.

1. If you subscribe to many mailing lists then learning to filter
your incoming mail is a must. If you use a shell account and then
learn the use of "procmail" or "maildrop" or "sieve". I believe
most webmail interfaces also offer some filtering.

2. Distinguish between spam and non-spam. Almost all mailing lists
receive spam and so having a spam filter in place is definitely
useful.

3. Filter incoming mail into separate folders based on the community
involved. One folder per mailing-list is not essential but is
sometimes the simplest thing to do!

4. Do not make the mistake of marking "noise" on the list as "spam".
You could effectively be cutting yourself off from the list if you do
so.  The "noise" usually has _specific_ characteristics that mark it as
unwelcome. Use your filter to select based on these characteristics
(given in increasing order of severity):
 a. thread (usually given by the "In-Reply-To" or other message ID
 headers)
 b. topic
 c. poster (this is the last resort!)

5. Be prepared to deal with a lot of incoming mail! Since most
mailing lists are archived you may wish to stop receiving mailing
list traffic (or auto-delete mailing list traffic) when you do not
have time to read the mailing list.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

ML Noise and Community (Was Re: [Ilugc] significance of deepavali)

[Kapil Hari Paranjape]

Hello,

A long response to Anthoni Shogun's query about why we don't just
ignore bad postings in this list.

FOSS is largely a _community_ activity. If people behave badly with
the community then even great ideas, coding or philosophy cannot
excuse them. If you don't believe me then see the criticism of people
like Stallman, Schilling and Shuttleworth over the last few years.

According to me, the "freedom" in FOSS is more in the sense of the
fundamental rights granted by the constitution --- these come along
with some fundamental duties. In summary, if you will not work with
the community then the community will not work for you.

Thus, a big part of working with FOSS is learning to use your ego
to drive you to be more creative rather than use it to fight other
creators. This involves learning to use mail, irc, comments, commit
messages and the like to _communicate_ on those things which other
people want to hear _in context_. A long comparision on the BSD
licence vs. GPL license will not be welcome within a comment section
of a source file even though it may be completely relevant in an
online forum.

There are two ways in which most community members react to stuff
that violates community guidelines. One is to ignore the poster.
This is the method used on high-volume lists. Some other lists like
ILUG-C have many more newcomers and are low-volume enough that
attempting-to-educate is a legitimate alternative. The latter may sound
like "noise" to those used to the former kind of list. However, note
that elimination of such noise from one's mail reader is just a matter
of learning to "kill-thread" in incoming mailing list mail.

Regards,

Kapil.
--
___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] significance of deepavali

[Kapil Hari Paranjape]

Hello,

On Sun, 18 Oct 2009, Anthoni Shogan wrote:
> If a person cannot top post to a Free and Open Source list, I really
  
> don't understand what is the kind of freedom we talk about all the
  ^^
> time.
  

That much is obvious! :->

Please read up on the topic and educate yourself!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] significance of deepavali

[Kapil Hari Paranjape]

Hello,

On Sat, 17 Oct 2009, Varrun Ramani wrote:
> Wish you all a very happy deepavali(not :) diwali )! I just was enlightened
> by my professor with some info  i had no idea about. I think every Indian
> ought to know this!
> 
> *Significance of Diwali*

I see ... and this is relevant to ILUG Chennai because?

 1. The "I" stands for India so everything about it is relevant?

 2. Because if you circulate this message then it will instantly
cause your knowledge about GNU/Linux to grow many fold!?

 3. The row of lights represents the successive versions of Linux and
other GNU software which will defeat the evil (but smart) Ravana
of proprietary software!?

I'm unconvinced. :-)

Having access to someone's mail address (or being subscribed to a
mailing list) does not mean that _everything_ of interest to you must
be circulated to all these addresses!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Downloading frame based sites with wget

[Kapil Hari Paranjape]

Hello,

On Fri, 16 Oct 2009, Varrun Ramani wrote:
> Does anyone know a hack/workaround/FOSS that would solve my problem?

Contact the web site maintainers and ask them for a copy? This is a
public institution after all!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (uniq - find non repeated lines in input)

[Kapil Hari Paranjape]

Hello,

On Fri, 16 Oct 2009, Roshan Mathews wrote:
> IIRC, another idiom was,
> 
> ... | sort | uniq -c | sort -nr | ...

Actually ' sort | uniq ' is so common that there is ' sort -u '

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Help With OpenVPN

[Kapil Hari Paranjape]

Hello,

On Mon, 12 Oct 2009, Raja Subramanian wrote:
> On Mon, Oct 12, 2009 at 7:52 PM, Hrishikesh Murali  wrote:
> In VPN parlance, what you are asking for is a full mesh VPN
> topology where any VPN site/network can communicate
> to any other site/network in the mesh directly.  Most commercial
> firewalls Cisco/Check Point/Juniper support this topology.
> 
> OpenVPN however only does star -- central VPN server with all
> clients communicating to the server only.

Thanks to Raja for figuring out that "mesh" toplogy is what
Hrishikesh wants.

As Raja has said OpenVPN only supports star structure (i.e. all
traffic goes through the VPN server) or point-to-point.

You can create a OpenVPN mesh network you need to to do a lot of
scripting to create many point-to-point networks and routing
tables will be long.

If the VPN structure is being used for encryption only and not for
tunneling, then a better alternative is to use IPSec or IPv6.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Help With OpenVPN

[Kapil Hari Paranjape]

Hello,

On Mon, 12 Oct 2009, Hrishikesh Murali wrote:
> I have the VPN server running successfully on 192.168.1.1 and the
> clients are connected to it (192.168.1.2 and 192.168.1.3).

If all the computers are on a common ethernet segment why are you
running a VPN between these machines in addition to the ethernet?
(If you want to encrypt data by default you are better of switching
all machines to IPSec.)

> Computer 1
> -
>   eth0 - 192.168.1.1/24 (This is the VPN server)
>   tap0 - 10.222.222.1

Note that tun devices are preferred to tap devices.

> Is there a better way to do the routing, that is can I push the
> routing entries 1) and 2) automatically from the VPN server onto the
> clients whenever they connect so that traffic does not have to come
> through the VPN server, it can go directly to the gateway?

Again your requirement is not very clear from your description. If
your VPN is meant to encrypt traffic then the traffic must pass
through the VPN server which behaves like the hub of a star network.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Vi-like browser (Was Re: [Ilugc] edit with vim)

[Kapil Hari Paranjape]

Hello,

On Sun, 04 Oct 2009, Manvendra Bhangui wrote:
> How about browsing the web with vim like interface :)
> 
> http://vimperator.org/trac/wiki/Vimperator

There are a number of alternatives. 

For "pure" text-mode joy of using vi try "elvis" which has a built-in
browser.

To get an a browser that is like vi in two senses (a) does only one
task --- web browsing (b) using vi-like "command mode" and insert
mode, try "uzbl" www.uzbl.org (pronounced usable I suppose!)

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Project idea/GSoC

[Kapil Hari Paranjape]

Hello,

On Sun, 04 Oct 2009, Mohan Sundaram wrote:
> The kind of annotations I'm talking of are typewriter inserts,
> rectangles, oval, circles, lines, arrows, comments, link boxes etc.

I don't know about all these things so my suggestion may be a bit
off. I _have_ done PDF markup in two ways:

 1. Write text "on top of" the PDF file. One can do this with flpsed
 which is short for "fltk PostScript Editor".

 2. Fill up PDF forms by extracting the FDF data. One can do this
 with pdftk.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Ubuntu Repository

[Kapil Hari Paranjape]

Hello,

On Sat, 03 Oct 2009, Kannan wrote:
> Like in Fedora, i want to copy all the dvd packages to my directory & want
> to use that repository using apt-get tool.

I use 'approx' (with Debian but it also works with Ubuntu). This
creates an entry in '/etc/inetd.conf' for a server running at .

You can now use the following lines (you must replace testing with the name of
your distro --- like jaunty or karmic) in /etc/apt/sources.list:

 deb http://localhost:/debian/ testing main contrib non-free
 deb http://localhost:/security testing/updates main contrib non-free

Finally, you can take all the debs that you have already downloaded
at import them with 'approx-import /path/to/downloaded/debs/*deb'

After this, you will not only get the debs you already have locally
but any updates will also be added to /var/cache/approx in a
directory structure like the one on the mirrors.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] An article bashing C

[Kapil Hari Paranjape]

Hello,

On Sat, 03 Oct 2009, Girish Venkatachalam wrote:
> http://aplawrence.com/Unixart/C_is_for_Crap.html

Even though the title is horrible, the article is worth reading for
the detailed information the author provides on what he considers are
the weaknesses of C.

Even though I/O has been pushed out to a C library we do not find many
people crafting their own libraries for these tasks. More importantly,
even those that do, stick to the "standard" interfaces.  This is
because "printf" and the like have been standardised.

There is no reason why proper memory management and exception handling
could not be provided by standardised C library interfaces.

There will always be programmers who will bypass the library and use
home-brewn alternatives but most people will use the standard way if
it is available (and hence taught in a first course on C).

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Is C always fast (Was Re: [Ilugc] .a-tip-a-day (patch - apply diffs and act as inverse of diff))

[Kapil Hari Paranjape]

Hello,

On Thu, 01 Oct 2009, Prem Kurian Philip wrote:
> From: Rahul Sundaram 
> 
> >Never say never. There has been quite a few demonstrations otherwise
> >available on the web. Feel free to look them up.
> 
> Again, it comes down to the algorithm used. If the algorithm used in C is
> poor, then it is likely to be slower than the same functionality
> implemented in a higher level language using a better algorithm.

Please do follow Rahul's suggestion and look it up!

Take the following program in LISP:

 (def fact (n) (fact1 n 1))

 (def fact1 (n k) (cond
((leq n 1) k)
(t (fact1 (sub 1 n) (times n k)

Let us translate this program/algorithm literally in C as:

 fact1(int n, int k) {
if (n <= 1) 
return k;
else
return fact1(--n,n*k);
 }

 fact(int n) { return fact1(n,1); }

It is quite likely[*] that the (compiled) LISP program will execute
faster and use less stack space than the C program. This is because
LISP compilers typically implement "tail recursion optimisation"
whereas C cannot (since C allows "side-effects").

> Since C is just a very thin abstraction over assembly language, it will
> tend to be quicker than pretty much anything else out there. Which also
> explains why the Linux kernel is implemented in it. The only thing quicker
> would be to get the compiler to output the code in assembly language and
> then hand optimize the code before running it through an assembler.

These "operating system related" things is what C is optimal
for. There are a _lot_ of different uses for computers and for many of
them C is _not_ optimal --- even in the sense of efficiency.

Indeed, all Turing complete languages _can_ implement each other's
features through translation. However, such translations typically
involve a small per-instruction cost which becomes significant cost
when a big project is implemented. Thus, each language has its own
"domain of expertise".

A thought for this thread:

 "A carpenter who only uses a swiss-army knife would be a poor
 carpenter, so why do some programmers insist on using only one
 language?"

Regards,

Kapil.

[*] I have qualified my claim about this particular program since it
is not large enough for differences to really show up.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

[Ilugc] Language metrics (Was Re: .a-tip-a-day patch etc.)

[Kapil Hari Paranjape]

Hello,

Since naming specific languages seems to activate the sensitive
spots, I will restrict myself to "abstract nonsense" here.

There are a number of terms that are used when measuring (programming)
languages. Sometimes these are called "language metrics":

 1. Popularity: The number of people reading/writing programs
in the language.
 2. Efficiency: The speed with which programs written in the
language can be run. (This might or might not include the time
taken to compile the programs).
 3. Readability: How easy it is to figure out what a program is doing
by reading the `bare' code. (Explanatory comments not included!)
 4. Writeability: How easy it is to convert a given algorithm into a
program in that language.
 5. High level/structured: How many different logical/mathematical/end-user
concepts are embedded in the language.

While there is doubtless a certain amount of subjectivity in assigning
values to these metrics, the real subjectivity in our judgement about
languages arises from how much weightage we assign to each of these.

 1. If we are writing for a specialised developer community,
popularity has little value.
 2. If we are writing small programs, efficiency of the running
code may not bother us at all. Also compiler efficiency is of little
relevance for daemons even if it is just-in-time compilation.
 3. If we do not really care whether others read our programs (e.g.
throwaway code) then readability is not an issue.
 4. If a large library of efficient builtin functions is provided, we
don't mind having to look up an index to use these functions. So
implementing algorithms is not "our job"; gluing together these
functions _is_.
 5. If we are manipulating very simple data objects in a
straightforward manner (such as reading/writing from devices) then
high-level concepts may even be a hinderance!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day (patch - apply diffs and act as inverse of diff)

[Kapil Hari Paranjape]

Hello,

I apologise for offending people and igniting a flamewar.
However, I could not prevent myself from writing the following long
response. :-(

On Wed, 30 Sep 2009, Manvendra Bhangui wrote:
> I am sorry about that. I honestly wanted to have some meaningful
> conversation and my point was only on someone making a comment that C is
> difficult to write and difficult to read - which is untrue if one knows the
> language.

While your intentions were honourable, your remark was about speed
and efficiency and number of syscalls --- none of which has anything
to do with reading/writing C!

> We all love the language which we are comfortable with. And a artist can
> take any tool and make a poetry or a painting out of it and not complain
> about the tool.

That is certainly true. There are indeed a large number of good
programs written in C --- however, even those programs would be
almost un-understandable if the comments are stripped out!

The kind of things you talk about, like assigning a char to an int
and so on are _precisely_ the things that make the language difficult
to understand since one has to run a small compiler + semantic
analyser in one's head to keep track of programs that do that.

C is a good language for writing operating systems, device drivers
and system libraries. That is one of the things that it was designed
for. Until Unix, most of the above tasks were programmed in
assembly. Using C allowed one to deal with them at a higher level and
create greater structure in the OS code, hence making it more readable
(than assembly code) and yet almost as efficient (as assembly) at the
same time.

Readable code is not necessarily efficient. For example, in Haskell,
one definition of the list of primes is:

  primes :: [Integer]
  primes = sieve [2..]
where sieve (p:xs) = p : sieve [x | x<-xs, x `mod` p /= 0]

This is almost exactly the mathematical definition of the Sieve of
Eratosthenes written thousands of years ago! So it is very readable
(comments are not required!). However, it is not an efficient program!

Now, scripting languages like Perl have some higher level facilities
so that it is comparatively easy to translate the above program into
it (retaining most of the inefficiency!). In C you would have to
implement list operations (and probably garbage collection) before
you could translate the program. However, in the course of doing
this you would find ways to make this more efficient as well! You
could also translate this new C program _back_ into Perl --- but
translating it back to Haskell would be significantly harder.

This example (if you work it out) will explain why I said that
programs are easier to write in Perl and easier to read in Haskell;
whereas neither is the case with C. As _you_ pointed out, good
programs in C do tend to be more efficient than any of these
languages and if a thoughtful author has provided enough comments,
then they become readable as well.

Hope this clarifies --- and douses the flames a bit!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day (patch - apply diffs and act as inverse of diff)

[Kapil Hari Paranjape]

Hello,

On Wed, 30 Sep 2009, Girish Venkatachalam wrote:
> Of course, that is a good thing. The utility is powerful but I
> don't like it. It should have been way more simpler and friendlier
> and still powerful.

Write such a (simpler and user-friendly) utility or at least design
one!

People criticise things like patch and Perl (and the English language)
for very sound _theoretical_ reasons. However, these are successful
systems and it is worth trying to understand why.

The "There's more than one way to do it" philosophy that underlies
these systems is one of the reasons people find it easy to adapt them
to their way of thinking and _create_ programs (or simple sentences
in English).

Systems like Haskell and Lisp (or languages like Sanskrit) are very
nicely structured and easier to read and understand once you learn
the basics. However, these are much more difficult to get started with
writing.

C is neither easy to write nor easy to read!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day (diff - differences between text files line by line)

[Kapil Hari Paranjape]

Hello,

On Tue, 29 Sep 2009, Girish Venkatachalam wrote:
> $ diff file.txt file-modified.txt

Generally, one should prefer the command
 $ diff -u file.txt file-modified.txt

The reason is that this maintains the _context_ information of the
differences as a result:
 - the diff is more human readable
 - the diff is longer :-(
 - the diff can work with patch even when some unrelated changes
   have been made to file.txt after the diff was created

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Fwd: Proprietary Tender by NIC for SC

[Kapil Hari Paranjape]

Hello,

On Fri, 25 Sep 2009, Roshan Mathews wrote:
> Hence, if they want Macs, then that's what they get.  If that choice
> means that others are forced to buy Macs too, then you complain, else
> it's just none of your business.

Imagine the furor if they put out a tender saying that they only want
to be supplied mobile phones made by Samsung or want internet service
only from Tata Indicomm.

Normally, a tender document should put out the specifications --- not
a commercial brand that is supplied by only one company. Government
departments over the years have found ways to subvert this system. In
this particular case, they will say that a number of different
"vendors" exist.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Regarding mail and chat clients (Was Re: [Ilugc] SFD Photos NEW!!] )

[Kapil Hari Paranjape]

Hello,

On Thu, 24 Sep 2009, pavithran wrote:
> You guys wont believe I also prefer using my own id "p...@pavithran.org"
> Its just a matter of mail client  which is stopping me from using it .

Perhaps you meant this posting as a joke but then I didn't get it!
:-|

> I prefer a web based mail client and I dont like squirrel mail ,horde
> or bluemail.

This sounds quite similar to "I would like to create doc files but I
don't like oo or abiword so I am stuck with MicroSoft". 

> I also want a mail client which could support me to chat with jabber
> protocol .

Why on earth would you want that? Mail and chat are completely
different protocols with very different goals. "I want a telephone
which also acts as a postal system."

> Yes I could settle for plain mail client but I believe in thin
> clients hence don't want a single mail to come to the client
> machine.

This makes little sense too. If the mail (at least the text of it)
"does not come" to the thin client how can you read it? If you mean
that the mail should be stored on the server then use a client that
supports imap and put imap on the server.

So really all that you need is a server that runs daemons that
support smtp, imap and jabber. Then a thin client that runs a mail
client and a jabber client is all you need. So for the server
something like postfix, dovecot, jabberd and for the client something
like mutt and irssi. (But you may prefer graphical clients like
balsa and pidgin).

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (figlet - create ASCII art text)

[Kapil Hari Paranjape]

Hello,

On Thu, 24 Sep 2009, Girish Venkatachalam wrote:
> How do you draw a cat?
> 
> Take a pencil and draw on a paper.
> 
> How do you draw a chair?
> 
> You pull it and sit down.

:-) Nice!

> Similarly you draw ASCII art with figlet. But only
>  ASCII characters, not arbitrary lines and drawings.
> 
> $ figlet hello

Some Debian folks declared figlet to be "non-DFSG" in 2004-5 as explained
here: http://bugs.debian.org/274950

There is a "DFSG-free" alternative to figlet with the unfortunate
name "toilet" that is available from http://libcaca.zoy.org/download.html

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] FOSS equivalent tool for the Teamviewer

[Kapil Hari Paranjape]

Hello,

On Wed, 23 Sep 2009, Ravi Jaya wrote:
> *The Problem Statement. *
> The primary objective is the desktop sharing, I need to share/access the
> LInux Desktop from the Windows as well as from the Linux box. The source and
> destination systems are wired across the Internet.
> 
> Any pointer or suggestion are highly appreciated.

The usual solution for this is "vnc". There is also a java applet
that allows you to view the desktop in a browser.

A different solution that has recently seen renewed interest is NX or
rather freenx.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Facinating songs about Linux?

[Kapil Hari Paranjape]

On Wed, 23 Sep 2009, Senthil Anand wrote:
>   I want audio and video songs that tells about the pride of
>linux,FOSS or opensource.

Since Unix was not about audio or video but about text (!), here is a
nursery rhyme from the time when Unix was a small child:

 http://www.mit.edu/people/yandros/humor/unix-alphabet

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (bc - interactive calculator)

[Kapil Hari Paranjape]

Hello,

On Wed, 23 Sep 2009, Girish Venkatachalam wrote:
> On Wed, Sep 23, 2009 at 6:30 AM, Kapil Hari Paranjape  
> wrote:
> > Hello,
> >  $ bc -l
> >
> > which is required if you want to use things like sine, cosine, etc.
> >

> This is driving me mad Kapil.

Sorry! ;-> I agree that the math library is a bit limited. More
complicated functions need to be defined and made available.

> But I have a problem. How will I convert from degrees to radians?

You can define a function in bc.

 define dtor (d) { return d*a(1)/45; }

The reason this works is that a() is the arctan or inverse of tan
function and tan(45 degrees)=1.

> But how to ask for sin(30), sin(60), sin(90) in degrees to radians?

 define sin (d) { return s(dtor(d)); }

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. (bc - interactive calculator)

[Kapil Hari Paranjape]

Hello,

On Wed, 23 Sep 2009, Girish Venkatachalam wrote:
> $ bc

Also note:

 $ bc -l

which is required if you want to use things like sine, cosine, etc.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. xargs command

[Kapil Hari Paranjape]

Hello,

On Tue, 22 Sep 2009, Roshan Mathews wrote:
> This is interesting.

Indeed.

> $ find -name "*.h" | xargs nano
> Received SIGHUP or SIGTERM
> 

> $ find -name "*.h" | xargs emacs -nw
> emacs: standard input is not a tty
> 

> $ find -name "*.h" | xargs vi
> Vim: Warning: Input is not from a terminal
> 104 files to edit
> 

> The issue does seem to be, as Kapil Paranjape pointed out,
> that vi has it's stdin set to /dev/null which it doesn't like.

It is not just "vi" the same happens to "emacs -nw" and "nano" which is
why the latter two exit. However, (like many unix commands) "vi"
assumes that you "know what you are doing" and starts up anyway,
whereas "emacs" and "nano" just quit since they "know what you mean".

> $ find -name "*.h" | xargs emacs
> 

> $ find -name "*.h" | xargs gvim
> 104 files to edit
> 

Since these are commands start a new window, they create a new
handler for input and so are not affected by setting stdin to
/dev/null.

Ideally, if one wants to use this method then a more appropriate
choice of command is "emacs-client" or whatever it is called
nowadays. This prevents yet another instance of emacs from being
started if one is already running.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. xargs command

[Kapil Hari Paranjape]

Hello,

On Mon, 21 Sep 2009, Roshan Mathews wrote:
> On Mon, Sep 21, 2009 at 10:19 AM, Kapil Hari Paranjape
> > I got the same results which leads me to suspect that the original
> > tip is faulty.

> The tip may not be faulty.

Perhaps you should try it and then make a pronouncement!

You could do this in a directory with a few files:
 $ mkdir /tmp/faultyfindq; cd /tmp/faultyfindq
 $ touch a{1..4}
 $ find . -type f | xargs nano

If you manage to do something sensible with the last command, then
please report it. (I have replaced vi with nano since Roshan seems to
dislike vi).

I suspect the problem is explained in this snippet from the man page
for xargs:

 --arg-file=file
 -a file
   Read items from file instead of standard input.  If you use
   this option, stdin remains unchanged when commands are run.
   Otherwise, stdin is redirected from /dev/null.

So the problem is that vi is run with stdin redirected from /dev/null
which will obviously not work.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] .a-tip-a-day. xargs command

[Kapil Hari Paranjape]

Hello,

On Mon, 21 Sep 2009, Venkatraman S wrote:
> On Sat, Sep 19, 2009 at 5:04 AM, Girish Venkatachalam <
> girishvenkatacha...@gmail.com> wrote:
> 
> > is a simple way to edit all files under /etc recursively including
> > directories. This is better.
> >
> > $ find /etc -type f | xargs vi


> Also, when i tried it with 5 files(txt), my cmd line got screwed up - as in,
> the keymaps werent working - up, down arrows were not responding. (is this
> only me or reproducible?)

I got the same results which leads me to suspect that the original
tip is faulty.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Freedom Bucket : A Bucket for Open Standards

[Kapil Hari Paranjape]

Hello,

On Sun, 20 Sep 2009, narendra sisodiya wrote:
> Freedom Bucket : is a idea coined by me to make a special directory
> in Linux based system which will contain on Open Format files like png for
> images, ogv , ogg for video and audio.

The idea is nice. Here are some remarks:
 1. Before trying for the daemon try to create a program!
 2. The conversion should be loss-less (this is difficult!). For
 example, conversion of bitmap to png or xpm is not lossy but
 conversion to jpeg is.
 3. After the program is working you can try to augment it using
 FUSE.
 4. There could be the option of not doing the conversion if it is
 not possible --- in this case there are two choices (left to the
 user), either not provide any file at all or provide the original
 file unchanged.

>   The aim of this project to remove non-standard and royalty based
> format from earth. It will be useful for those persons who want to use only
> open formats.

One of the additional difficulties is that some formats are "open"
but are patent encumbered. Trying to maintain "purity" in this
context may be difficult.

Best regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Hacker Ethics (Was Re: [Ilugc] I will take care)

[Kapil Hari Paranjape]

Hello,

On Sat, 19 Sep 2009, Girish Venkatachalam wrote:
> In fact if you ignore the rebirth theory, then there is really no
> motivation for people to be ethical and moral...but I digress as
> usual. ;)

I think that many hackers go by the ethics of "Enlightened
Self-Interest".
http://en.wikipedia.org/wiki/Enlightened_self-interest.

On the other hand one must take care not to become too learned
otherwise one may not be reborn! http://en.wikipedia.org/wiki/Bodhi

:-)

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] query about virus attack on UNIX / LINUX environment

[Kapil Hari Paranjape]

Hello,

On Thu, 17 Sep 2009, Girish Venkatachalam wrote:
> Cryptography requires randomness in everything. Crypto alone cannot
> solve security problems.

I will take a different fork here from what Girish said and add
that programs alone cannot ensure security for a computer system. A
good system _encourages_ good security practices. Some examples of
such practices that are the _default_ on most *nix like systems.

 1. Ensuring that people do not use the administrator account
 for day-to-day work.
 2. Providing means to check that software is downloaded from reliable
 sources.
 3. Providing enough software from reliable sources at low cost so
 that people are not encouraged to depend on unreliable sources.
 4. Clearly demarcating system areas from user areas so that users
 can ensure the integrity of these areas _independently_.
 5. Using privilege separation to limit access to files,
 executables and devices.
 6. Providing a path by which users can learn more about their system
 in order to improve their security through better understanding.
 7. Educating the user through (a) clear notifications when they attempt
 something potentially risky and (b) explicit documentation of the
 vulnerable aspects of their system.

I am sure I have missed some, but you probably get the idea. The point
is that most of these practices are _less_ convenient on Windows than
they are on a generic *nix. Security is always fighting with
convenience so reducing the inconvenience of secure operations is
always important.

Effectively, Windows systems encourage their users to use them in a
manner which is not secure![*]

Regards,

Kapil.

[*] One of my biggests fear is that modern "easy-to-use"
Desktop versions of GNU/Linux will copy this unfortunate
"ease-of-use-wins-over-security" attitude of Windows.

--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] One Day One GNU/Linux Command (DOS2UNIX) ((tag: odoc, linux))

[Kapil Hari Paranjape]

Hello,

On Thu, 17 Sep 2009, Girish Venkatachalam wrote:
> In the shell do this.
> 
> $ jot 50 1 50

An alternative for "bash" users is
 echo {1..20}
This can be taken further with things like
 echo file{1..20..2}

And so on.

Look at the section on "Brace Expansion" in the "bash" man page for
more information.

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] total failure to find a netbook with linux

[Kapil Hari Paranjape]

Hello,

On Wed, 16 Sep 2009, Mehul Ved wrote:
> Would it really be the case? From what I am aware, most manufacturers
> will load the Windows machines with trial software from *reputed*
> vendors, which lowers the cost of the machine setting back some of the
> higher cost. So, now the machine isn't priced too high as compared to
> the Linux/DOS offering.

This may be the case for machines that cost 30K and more. Most of
this trial software is games etc. that are not very likely to be
activated by those who buy netbooks.

When the MRP of netbooks _could_ have dropped to about 15K or even
less (for example, see the Norhtec page referred to earlier), the
manufacturers have added "features" to raise the price to the 25K-30K
range. One of these features is the inclusion of Win XP which by all
accounts they are getting from MS at about $20 or less.

Now, all of these taken individually seem like common business
practices:
 1. Not selling ultra-cheap goods. (OEM's)
 2. Adding features to raise prices and margins. (OEM's)
 3. Under-pricing a value addition so that it can be bundled. (MS)

However, put it all together and it feels a monopolistic trade
practice designed to keep competing products out. At the same time
there is a business opportunity for someone to come in and sell
significantly cheaper computers exactly as Asus did about two years
ago.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] total failure to find a netbook with linux

[Kapil Hari Paranjape]

Hello,

On Tue, 15 Sep 2009, varadarajan narayanan wrote:
> May be you may find this interesting !
> 
> http://playingwithsid.blogspot.com/2009/06/netbook-for-developing-countries.html

Someone else also had a similar idea. Check this out
 http://www.norhtec.com/products/gecko/index.html

In so many ways this may be the perfect netbook. Wonder why it does
not have more publicity?

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] wanna being with lxde

[Kapil Hari Paranjape]

Hello,

On Sun, 13 Sep 2009, Rahul Sundaram wrote:
> On 09/13/2009 07:11 AM, Kapil Hari Paranjape wrote:
> > On "ubuntu" this is not easy to do. Many "standard" system services are
> > tied to gnome. So the basic "ubuntu-desktop" pseudo-package depends
> > on a number of gnome/gtk packages.
> 
> So, what's the problem with that? Meta packages are for convenience.
> Removing them is not going to break anything.

That is true. I should have said that the packages that
"ubuntu-desktop" depends on provide a number of services that a
"standard" desktop depends on. So, if you want to remove all these
(gnome-based) packages then you probably need to replace them with
others that provide similar functionality.

In the case of kubuntu, some kde-based alternatives are provided
while xubuntu uses some gtk-based alternatives. I don't know what
the replacements for "lxde" will be but perhaps "lubuntu" will make
appropriate choices.

On the other hand you may not want some of the functions provided by
these packages and are willing to organise your desktop/workspace
etc. on your own.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] wanna being with lxde

[Kapil Hari Paranjape]

Hello,

On Sat, 12 Sep 2009, NavTux wrote:
> how to remove my gnome to work with lxde only?
> is it possible to remove my gnome?

On "ubuntu" this is not easy to do. Many "standard" system services are
tied to gnome. So the basic "ubuntu-desktop" pseudo-package depends
on a number of gnome/gtk packages.

To reduce this dependence people created "kubuntu" and "xubuntu". The
most recent project is "lubuntu" which is probably what you want. As
far as I know the last project is still under development. You may
want to join in to test and develope "lubuntu".

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] How to find the OS installed date?

[Kapil Hari Paranjape]

Hello,

On Sun, 13 Sep 2009, Rajesh Pandian M wrote:
> p.s:there would be a cleaner way than this definitely :-)

If you used the Debian installer while installing the system (which
is the usual way of installing a fresh system) then you would have 
log files in /var/log/installer/ that provide detailed information
about what the installer did along with time stamps.

However, note that these time stamps are usually based on your
system's hardware clock so if that is not accurate, then the time you
get will not be.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] total failure to find a netbook with linux

[Kapil Hari Paranjape]

Hello,

On Sat, 12 Sep 2009, Vamsee Kanakala wrote:
> In consolation, you might at least have a good light-weight os to
> install: http://bit.ly/rZX2q. easy peasy looks cool, and I love
> their homepage :)

This looks very much like a re-theming of Ubuntu Netbook Remix.

It is not very clear from their wiki/home page what exactly they have
changed. (I admit that I have not researched this extensively!)

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Restrict users commands

[Kapil Hari Paranjape]

Hello,

On Wed, 09 Sep 2009, Govi wrote:
> I want to restrict users normal command.i mean specified users must use only
> below command,

1. Make rbash the user's shell using change shell.

2. Set the PATH to something like /usr/local/rbash/bin

echo "PATH=/usr/local/rbash/bin" >> /home/user/.profile
chown root.root /home/user/.profile
chmod og-w /home/user/.profile
chattr +i /home/user/.profile

3. Link the binaries to this directory

for i in mv dig telnet ssh
do 
ln -s $(type -path $i) /usr/local/rbash/bin/
done
chown root.root /usr/local/rbash/bin/

4. Test it out! (Perhaps you need to copy or hard link the binaries
and symbolic links are not allowed).

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Re: USB bootable Linux distribution (Lokeswara Bheeman)

[Kapil Hari Paranjape]

Hello,

On Wed, 09 Sep 2009, Varrun Ramani wrote:
> The Preferred USB bootable GNU/Lunix distro would be Ubuntu. Download the
> latest version Jaunty Jackalope, boot the live CD, plug in your USB. Then,
> go to System->Administration->USB Startup Disk creator. Select the image of
> Jaunty on your hard disk and it automatically creates the USB Startup disk.

An alternative is to use unetbootin from "any" operating system.
The unetbootin does not require booting into the live CD.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Obama and Linux

[Kapil Hari Paranjape]

Hello,

On Tue, 08 Sep 2009, Aanjhan R wrote:
> "Tyrants" - Google gives me the following synonym (a cruel and
> oppressive dictator )

While I do not approve of tit-for-tat, consider how the proprietary
software industry throws around the word "pirate"!

> Imagine he speaking on them creating the next "linux" kernel. Do you
> really think the majority of kids will get inspired and their
> creativity circuits in their brain triggered by that?

Unfortunately, you are right. They would also not recognise many
important scientists. Most school kids probably think of Bill Gates as
a computer scientist and would say "Alan Turing? Who?"

There is indeed a fine line between a leader leading so that his
followers can follow and a so-called "leader" who just says what he
thinks his "people" want to hear. In this case, Obama is trying to
underline the importance of innovation (scientists/engineers) as a
career over the usual management/sales jobs. To my mind that is
a good thing. The choice of example (iPhone/Steve Jobs) is not as
relevant as long as it conveys the message.

As long as his government is aware of this distinction and does not
support the proprietary software industry in its anti-innovative
practices, we should not complain too much.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Issue Installing Fedora 11

[Kapil Hari Paranjape]

Hello,

On Fri, 04 Sep 2009, Amit Dey wrote:
> sd 2:0:0:0: [sda] Attached SCSI disk
> 
> sd 2:0:0:0: Attached scsi generic sg1 type 0
> 
> ata4: SATA link down (SStatus 0 SControl 300)
> 
> atkbd.c Unknown key pressed ( translated set 2, code 0x29 on isa 0060/serio0)
> 
> atkbd.c Use 'setkekycodes e059 ' to make it known.
> 
> atkbd.c Unknown kkeky released( translated set 2, code 0x29 on isa 
> 0060/serio0)
> 
> atkbd.c Use 'setkekycodes e059 ' to make it known.

The messages about the keycodes do not seem to be the likely cause
of the problem. On the other hand the message about the SATA link
down seems more serious.

> Hope I have clearly put out the problem this time :)

Much better!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Key Bindings

[Kapil Hari Paranjape]

Hello,

On Thu, 03 Sep 2009, Girish Venkatachalam wrote:
> On Thu, Sep 3, 2009 at 10:17 AM, Kapil Hari Paranjape 
> wrote:
> > Overall, I have found that messing around with loadkeys is only
> > useful if you _only_ use the Linux console. Even in that case it is
> > better to define key sequences in specific applications rather than
> > replacing the characters that the console stuffs into the buffer.
> 
> Sorry but I beg to differ.
> 
> I even think that you could be wrong here. We are only replacing
> the string buffer , and in this case for F1 it was a totally
> useless representation. We are not really diddling with the
> keycodes or anything like that.

The problem is that you may accidentally hit "F1" while in application
"foo" for which "ls -l\n" is a command that deletes all your files! :-(

(If you have read the book "Hitchhiker's guide to the galaxy" then
there is some point where it warns that "Careless words can cost
lives"; this is a similar situation).

One way to make it work for bash regardless of which terminal you open
(console, xterm, gnome-terminal, etc.) is long as it generates the
same sequence "ESC+[OP" for F1 is to configure this in ~\.inputrc as
follows:

 $if bash
 "\e[OP": "ls -l\n"
 $endif

This definition will not mess with _any_ other application.

> > Most terminal applications (vi,emacs,mc,git,screen ...) and a few
> > graphical ones (like Emacs-with-X, gvim) allow you to assign commands
> > or character strings to key sequences.
> 
> Not useful I think. As I said, the use case is different.
> 
> I don't even want to switch on the monitor.

This is one use case for which there is no alternative to loadkeys.

However, you should be aware of Consequences (note the big C). For
example, the Alt+sysrq+r combination is directly grabbed by the kernel
and you can _never_ use it in another application as long as the sysrq
feature is enabled (via proc) --- and it "does" have disastrous
consequences if you happen to be using X or vi at that time.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Key Bindings

[Kapil Hari Paranjape]

Hello,

On Thu, 03 Sep 2009, Girish Venkatachalam wrote:
> Set it to
> 
> string F1 = "\033[[ls -l\n"

This looks wrong. I think it should be
 string F1 = "ls -l\n"
The additional Esc+[+[ seems to be unneccesary.

> Obviously this done not work in xterms. I never got that to work.

The above key sequence settings may cause additional problems with X.
For example, did you use "xev" to check whether pressing "F1" still
gives the "F1" XKSymbol after this wierd definition?

Overall, I have found that messing around with loadkeys is only
useful if you _only_ use the Linux console. Even in that case it is
better to define key sequences in specific applications rather than
replacing the characters that the console stuffs into the buffer.

Most terminal applications (vi,emacs,mc,git,screen ...) and a few
graphical ones (like Emacs-with-X, gvim) allow you to assign commands
or character strings to key sequences.

The later in the food chain that you assign a key sequence, the more
likely it is that the meaning (semantics) can be application specific.

It is true that it would be nice if there was a general way to say
something like the following for all applications at once:
 F1 should be help
 F2 should be prompt for command/function
 F3 should be prompt for external file/URL 
and so on. Moreover, this should be user configurable in one place
for all applications.

Clearly this needs some kind of hotkey callback library that all
applications should link to. Perhaps the Desktop Environments do this
but I wouldn't know since I don't really use them.

Regards,

Kapil.
--
___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] netbook recommendations

[Kapil Hari Paranjape]

Hello,

On Tue, 01 Sep 2009, Immanuel wrote:
> My experience:

> Overall I find HCL system a good buy with linux for me.

Could you please mention the precise model and specs? If you can
specify in which Tata Chroma you got it (which city), that would be
nice too! It sounds like a very nice netbook from your description.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] netbook recommendations

[Kapil Hari Paranjape]

Hello,

On Mon, 31 Aug 2009, varadarajan narayanan wrote:
> And distro makers can not support all the available netbook /laptop
> models. So the end user is left to tinker with the linux
> netbooks/laptops  on their own and depend on forums or their local
> linux guru !

This is not totally true at the moment. As far as I know, Ubuntu
Netbook Remix and Moblin work on a number of different Netbooks
(at least Acer Aspire One, Lenovo S9/10 and MSI Wind) without any
serious tinkering --- just follow the instructions on the web site(s).

One of my colleagues even installed MacOSX on the MSI Wind --- but
that _was_ serious tinkering!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] netbook recommendations

[Kapil Hari Paranjape]

Hello,

On Mon, 31 Aug 2009, Kenneth Gonsalves wrote:
> what about this:
> http://www.techgadgets.in/laptop/2008/04/acer-aspire-one-laptop-offers-mobile-
> internet-solution-in-india/

The date of the entry is the problem.

  "This entry was posted on Wednesday, June 4th, 2008 at 7:12 am and is
   filed under Laptop."

This is the netbook A150 that I have and bought in November 2008.

In early January Acer announced that they were dropping the price by
Rs. 3000/-. Shortly after that the new model D250 was introduced and
the old one went "out of stock"; the new model comes with Windows
pre-installed.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] netbook recommendations

[Kapil Hari Paranjape]

Hello,

On Mon, 31 Aug 2009, Kenneth Gonsalves wrote:
> I want to buy a netbook - cheapest possible. Any recommendations for a linux 
> compatible one, and where is the best place in Chennai to buy from?

Almost any netbook that is available in Chennai today:
 1. can run with Ubuntu Netbook Remix
 2. has Windows pre-installed

The counter-examples to (1) are the older Celeron-based models
that should be avoided since these are battery hogs and are slower
to boot than the Atom powered ones.

If you find one that is a counter-example to (2) then I guess
everyone on the list would like to hear about it!

People keep saying that the Arm-based netbooks will be released "any
time now" but I have yet to see one in the market or available in
India online.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] face as password

[Kapil Hari Paranjape]

Hello,

On Mon, 31 Aug 2009, Prem Kurian Philip wrote:
> If your entire ID mechanism relies on this compromised biometric code,
> then the person will have to be locked out of the system - because the
> person cannot change their finger / iris print once they become public
> knowledge.

Bruce Schneier once said something like:

 You can change your password but you cannot change your fingerprint.
 Think about it. Then think about it once more!

It it slowly becoming commonplace to use multiple independent sources
of authentication on computers just as we are supposed to in "real
life". The independence of the sources is important of course. So it
seems likely that authentication should be based on something which
you have _and_ something which you know.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] learning C++

[Kapil Hari Paranjape]

Hello,

On Fri, 28 Aug 2009, Kenneth Gonsalves wrote:
> my daughter is doing state board 12th - she has C++.

I thought you would have taught her some Python by now! So she would
only need to "translate" from Python to C++ and then check for syntax
errors.

Are you sure that they are teaching standard C++? Often they teach
using one of the MS Visual C++ type systems and that is quite
different from using g++.

Otherwise, the tools already suggested sound nice.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Webcam

[Kapil Hari Paranjape]

On Thu, 27 Aug 2009, Rajesh kumar wrote:
> Please Tell me some of the tools to take photo from laptop webcam

The following are for USB Webcams.

uvccapture
luvcview
guvcview

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Open Source Backup Software

[Kapil Hari Paranjape]

Hello,

On Thu, 13 Aug 2009, Girish Venkatachalam wrote:
> You can find a lot more backup URLs in my delicious page.
> 
> http://delicious.com/girish1729/backup

You seem to have missed "duplicity" which has some nice aspects:
 1. It uses cached metadata to decide what to backup.
 2. It saves things as compressed tar so that you do not run
into file inode issues.
 3. You can encrypt backups. So backups to remote machines are safe.

The main drawback of duplicity is that the _first_ run (where it
builds the metadata tables) takes a long time. This is also the
drawback of the other rsync based systems.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] a new website

[Kapil Hari Paranjape]

Hello,

On Tue, 11 Aug 2009, Kenneth Gonsalves wrote:
> http://ilugc.org.in

Could you please enable login using OpenID? 

On Tue, 11 Aug 2009, Bharathi Subramanian wrote:
> Check the ILUGC.in site demo also:
> See http://demo.ilugc.in/

This site seems to have a little more information than the previous
one.

On Tue, 11 Aug 2009, Roshan Mathews wrote:
> There are also http://www.ilugc.com/ http://www.ilugc.net/ and
> http://www.ilugc.org/

Since ILUG-C is not a commercial organisation I would prefer that the
www.ilugc.com contains a permanent redirect to one of the other
sites. A .net site should really be a network service site which
ILUG-C is not so I think that should be a permanent redirect too.

I don't know about others but I would also prefer a .in registration
for the "default" home page.

All the same ILUG-C is an organisation of sorts so a .org domain is
probably OK as well.

People, please co-ordinate and enable co-operative management of
these sites rather than a proliferation of web sites!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] How to generate SSLv3 certificate using OpenSSL

[Kapil Hari Paranjape]

Hello,

On Tue, 11 Aug 2009, Rosario Arun wrote:
> $openssl ca -policy policy_anything -config ./openssl.cnf -cert
> certs/ca.cer -in ./requests/certreq1.txt -keyfile ./keys/ca.key
> -days 360 -out ./certs/iis1.cer
> 
> After creating the certificate, i found its not SSLv3. How do i
> force the above command to create a certificate SSLv3. ?

If your "openssl.cnf" is the standard configuration file that comes
with openssl then using "-extensions v3" as an additional option
should be enough.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] 40th anniversary of Unix approaches...

[Kapil Hari Paranjape]

Hello,

On Mon, 10 Aug 2009, Mohan Sundaram wrote:
> I'd written a mail on this earlier in a thread where NASSCOM rejecting
> FOSS standards was discussed. I see the reasons as being less
> sociological than pragmatic. My views are as under:

The reasons you have given are reasonable for:
 (a) Organisations/people who were newbies in the 90's
 (b) Business organisations/people (who _must_ follow the market).

This still does not explain why the organisations that had competence
in Unix allowed this competence to lapse (for example, by "forgetting"
to pass it on to the younger generation). Moreover, it looks like
academic organisations are following business trends rather than
leading businesses to future trends.

> It is still not too late.

One always hopes that it is never too late! :-)

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

[Ilugc] Projects in FOSS --- a world wide problem?!

[Kapil Hari Paranjape]

Hello,

Since a recent long thread here was on how difficult it is to get
departments to approve FOSS projects here is are some links:

 


 

So at least the problem is a "World-Wide Phenomenon"!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] 40th anniversary of Unix approaches...

[Kapil Hari Paranjape]

Hello,

On Sun, 09 Aug 2009, Sivasankar Chander wrote:
>   There's an article in today's Hindu on the 40th
> anniversary of Unix. It has some inaccuracies, but is still a nice
> read overall:
> 
> http://www.hindu.com/2009/08/09/stories/2009080952761000.htm
> 
> Shameless plug: it contains some quotes from my advisor,
> Prof.Gopinath at IISc, Bangalore.

One point that is worth noting is that Indian academic institutes and
banks in the late 80's and early 90's got their systems from Indian
software organisations which specialised in Unix.

An interesting thesis topic for a sociologist is how this "Unix
country" became a "MicroSoft country" in the late 90's and the early
21st century.

At least part of the reason was that these organisations looked upon
Linux and GNU as "toys" for "hobbyists"---not for professionals. They
failed to recognise that Unix itself had similar roots.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

[Ilugc] Leaving Chennai

[Kapil Hari Paranjape]

Hello ILUG-C Folks,

I am shifting to Chandigarh for the next two years to teach at IISER,
Mohali. So you probably won't be seeing me at the meetings for a
while. I will still lurk around on this list.

Being part of ILUG-C activities and discussions has been fun for the
last few years --- it is unfortunate for me that I joined in so late.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] how to log in to a remote server with ssh2

[Kapil Hari Paranjape]

Hello,

On Fri, 07 Aug 2009, Kenneth Gonsalves wrote:
> I have libssh2 installed, but how do I log in to a remote server using it? I 
> tried typing ssh2 on the terminal and it did not recognise the command. It 
> sounds silly, but google did not give me an answer.

"libssh2" is a library for using the "Secure Shell Protocol" (version
2). You need to install a program that uses this library for using it
to connect. For example, I think "libcurl3" uses the "libssh2"
library and "curl" uses the "libcurl3" library.

So you can use "curl" to make a secure shell connection using this
library. The following two examples are from the "curl" manual:

  Get a file from an SSH server using SFTP:

curl -u username sftp://shell.example.com/etc/issue

  Get a file from an SSH server using SCP using a private key to authenticate:

curl -u username: --key ~/.ssh/id_dsa --pubkey ~/.ssh/id_dsa.pub \
scp://shell.example.com/~/personal.txt

To make a secure shell connection in the usual way you need a client
like "openssh-client" or "dropbear".

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] telnet port is safe fot SMTP purpose ?

[Kapil Hari Paranjape]

Hello,

On Fri, 07 Aug 2009, Ramkumar wrote:
> Telnet (23) is safe for smtp purpose instead of 25? give
> some idea abt it.

Any IP port is like any other so the choice of port 23 instead of
port 25 will not make mail any more or less safe. Safety will be
determined largely by the SMTP daemon you run and how well you
configure it.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Scheduled/Timed Suspend and Resume

[Kapil Hari Paranjape]

Hello,

On Wed, 05 Aug 2009, Roshan George wrote:
> The reason I'm asking is that I have a Pentium D computer which acts as
> a file-server/DVR. Now, as everyone (except me when I bought the
> motherboard that only supports these) knows, these are power-hungry
> monsters so I intend to keep it in sleep most of the time. I'm assuming
> I can get it to Wake-on-Lan, so serving files isn't a problem. However,
> I have no solution for scheduling it to wake up to record. Any
> suggestions?

Have some process run some "knock" port like trying to ssh to it to
get it to wake-on-lan.

The kernel is "asleep and cannot wake itself up". Someone/something
must interrupt the sleep from "outside" via the BIOS.

A different option is to use the power management features of the
kernel. Put the processor into low speed state and HDD to spin
down. This does not conserve as much power as a suspend-to-ram but may
be enough for you.

The kernel will spin up the drive and ratchet up the processor speed
as and when required. (BIOS does not play a direct role here). The
startup is also significantly faster.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

[Ilugc] Mirror at iit kanpur

[Kapil Hari Paranjape]

Hello,

The following information may be of interest.

http://mirror.cse.iitk.ac.in/

Official mirror of:

Arch
Fedora
Ubuntu.

The debian mirror is functioning but is not officially listed.

This info was received from Piyush Kurur of IIT Kanpur.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] how to connect two internet connections into one system?

[Kapil Hari Paranjape]

Hello,

On Thu, 30 Jul 2009, kesavan (K7) wrote:
> here using two internet connections by using two seperate systems
> but i need that two net connections are connected into same one
> system after that connections are sharing into client systems

You mean you want to use two connections simultaneously? You want to
look for "load balancing" network sharing on your favurite search
engine. Typically, for WAN links this is done by choosing one route
for each new outgoing TCP/UDP connection. You also use round-robin DNS
so that new incoming connections can choose a route "at random".

Once the gateway machine does this and also does network address
translation (NAT), the machines behind the gateway need not care how
the load balancing is carried out.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Debian adopts time based release

[Kapil Hari Paranjape]

Hello,

On Wed, 29 Jul 2009, Rahul Sundaram wrote:
> On 07/29/2009 10:31 AM, Sivanandhan, P. wrote:
> > Debian project adopts time based release
> > 
> > News Announcement http://www.debian.org/News/2009/20090729
> > 
> > Congratulations Debian team!.
> 
> I would note the subtle difference between a time based development
> freeze and a time based release in the actual announcement. They are not
> the same thing.

Indeed freeze and release are not the same thing. Occasionally long
freeze periods have led to many people to freeze their work on Debian
which is also a bad thing!

So I see this announcement with mixed feelings.

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] Gyaan on Partitions

[Kapil Hari Paranjape]

Hello,

On Fri, 24 Jul 2009, Venkatraman S wrote:
> What i am looking for:
> - How partitions work (the technical angle and insights)
> - Understanding Partitions (user's perspective)
> - Effective partitioning techniques or methodologies

On Fri, 24 Jul 2009, Kenneth Gonsalves wrote:
> -How to shrink a logical partition

It is no longer necessary to learn how to partition a disk in the
DOS sense.

Choice 1: Just use parted with your favourite interfaces,
command-line, curses, GTK, QT, ...?

Choice 2: Make just one LVM partition which you can then use via LVM2
in any way you like (parted also speaks LVM). Then use grub-pc (grub2)
to boot your system.  (Or if you insist you can use lilo as well).

One advantage of the second method is that you can resize your
logical block devices later as you find you need less/more space.
In this case the static "mbr" can be retired along with Windows!

Regards,

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with 
"unsubscribe  "
in the subject or body of the message.  
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc

Re: [Ilugc] MS contributing to Linux

[Kapil Hari Paranjape]

Hello,

On Tue, 21 Jul 2009, தங்கமணி அருண் wrote:
> Read : http://lkml.org/lkml/2009/7/20/167

This is indeed good news. At least GPLv2 is not a "virus" according
to _some_ people in MicroSoft.

Now, if only they submit some of their patents to OIN (open invention
network) so that:

 1. Mono/.Net and Moonlight/Silverlight can gain wider adoption.
 2. Vfat can actually be used on all portable devices.
 3. Document formats for office use can converge.

If they do this we might even find them more friendly than some other
computer giants.

Just like the Hyper-V driver provides compatability _and_ thus
improves the chances for people to continue to use Windows, so will
the above actions provide a net gain for MicroSoft.

Or perhaps I am dreaming!

Kapil.
--

___
To unsubscribe, email ilugc-requ...@ae.iitm.ac.in with
"unsubscribe  "
in the subject or body of the message.
http://www.ae.iitm.ac.in/mailman/listinfo/ilugc