Photo editing with Freedom – the GIMP way

 

Until recently, the issues with FOSS and photo post processing was loss of data while RAW conversion, poor colour management, lack of 16  bit support etc.

Finally it seems Gimp 2.9 is addressing the 16 – bit issue which was until recently addressed only by CinePaint. Yes there are multitude of options like Photivo, Krita etc but IMHO if all these efforts are combined in building a better GIMP, RAW support etc things could have been better. Looks like too much of a freedom is not always the right way unless there is a guiding force.

Gimp 2.9  30 May 2013 development branch

Gimp 2.9 30 May 2013 development branch with a unified / single window, support for images with upto 32 bit floating point precision (hurray!!), better colour profile management – trust me! my TIFF images looks the same in Mac OS X & GNU/Linux now, GEGL

Here I am trying to get Gimp 2.9 working on GNU/Linux flavours and if possible on Mac OS X as well & explaining photo processing with Gimp 2.9. I am thinking of using RawTherapee (http://rawtherapee.com/) for RAW conversion as I am not happy with UFRaw. For the colour management I am using Open-ICC profiles, Argyll CMS & UI

The workflow:

  1. Photos are taken in Adobe RGB 1998
  2. Monitor uses calibrated profile
  3. Processing is done with ProPhoto profile from OpenICC bundle
  4. RAWTherapee is used for RAW conversion
  5. Final output in JPEG and TIFF are saved with Adobe RGB 1998
  6. Gimp 2.9 compiled from source is used as it supports 16 – bit image formats
  7. RAM is used for GEGEL swap for faster performance by mounting tmpfs with 6GB (instead of default 50% physical memory size)
  8. Linux SWAP is also getting used upto 2.8 GB

 

Setting the stage

The first step is to install OpenICC profiles

Compile Gimp 2.9 as per the instructions

1. http://crunchbang.org/forums/viewtopic.php?id=25716

2. http://www.gimp.org/source/howtos/gimp-git-build.html

Install RAW therapee & get an image ready.

 

Errors & fixes:

The first error I came across was babl accessing a different process and can be fixed with a less “secure” kernel setting as shown below.

Change ptrace_scop by Editing  /etc/sysctl.d/10-ptrace.conf & the resultant value should be: kernel.yama.ptrace_scope = 0

Keel in mind that this will be applicable only after restart & we can change proc values as follows.

# echo 0 >  /proc/sys/kernel/yama/ptrace_scope

 

Second error:

Yet to find a solution for this.

babl-memory.c:114 babl_malloc()

(script-fu:1976): LibGimpBase-WARNING **: script-fu: gimp_wire_read(): error

 

 

I will be updating the document with more insights and automation. For now, the above steps should work and may be you want to look into some GIMP tutorials like: http://blog.patdavid.net/2013/03/the-open-source-portrait-postprocessing.html

 

Updated: 30-05-2013

SPAM. I hate it. So do you

Just found an old article I wrote on Spam. I had won first prize for an article competition for this one. I am quite proud about it 🙂 and I ended up writing this since someone had spoofed my email. I have no clue as to what happend to the prize / prize money I received ! Apparently the only thing I remember is finding about Domain Keys just few minutes before the final deadline of the competition. Oh yes, its written using LaTeX

So here it is: SPAM. I hate it. So do you 

Pigeons in the hostel and Google+

Now that Google has re-realized the potential of social networking or cashing on human relationships I am writing down my ramblings on the phenomenon called “everyone needs to satisfy their ego or attention” aka Facebook effect.

Somehow I was lured into the social networking arena long back I remember signing up with a alternate mail address in Orkut and filling in minimal information because of  privacy concerns. Irony was when in 2005 I became one of the first cloned human beings and had trouble chasing down the culprits. (Finally decided not take legal action as two of the involved parties will face deporting from Gulf and permanent ban). The positive side of the incident was I became increasingly aware of the potential and dangers of the medium, read about anything and everything in the topic including the definite papers from Orkut. One of the most interesting communities I found on Orkut was “I hate pigeons in the hostel” This was a perfect example the ideas explained in the paper “Exploiting geographical location information of web pages” but here, not just location information was shared, but a person’s innermost feelings and opinions about pigeons were available at the disposal for exploitation / targeting. Eventually I had moved out of social networks and tried to build a real world network called Koottam.org along the lines of certain couch sharing projects and is still pretty much active.

 

Let’s take deviation and take a look at Google’s revenue stream. Most of it comes from display advertisement even though the company is fighting Apple in the handheld market and Microsoft in the enterprise market by providing office clones via browsers. To complete the eco-system Google Appengine and enterprise market place also is being established. But all these businesses are still in pre-alpha mode compared to the display advertiment market which it effectively does by  making use of content partners aka Google Display Network, various frontends like Gmail, Google groups, Orkut, AdSense etc. Though poineers in social networking Google never seemed to be interested in the business until Facebook started pushing Like buttons and various share options aggressively. At the time of writing major tech portals only allow one to post comments after signing in via Facebook or Twitter and of the 5 random display partners from the Google display network I verified talks nothing about Google but has either Facbook Like or Share enabled and some even have a facebook fan page occupying half the home page. In short Facebook was slowly eating into the core business of Google and sharing this information with a new search engine called Bing. This new search engine had recently established a praetorship with Yahoo! another not so big search engine. Along with this, the internet killer, sorry, Facebook slowly started replacing content with Facebook pages and presents not so cool, but easy statics about their pages to every one  500 million people who was unware of online advertising or tools like Urchin (Google Analytics) or Google Adwords.

 

I am not sure how many of the google display partnersites are integrated with Facebook but considering the large number of penetration, including Youtube its pretty much evident that Facebook/Bing tracking systems gets updated about activity in Google’s vast display business simultaneously. And there was a big catch Facebook and thus Bing always knew whether the visitor likes pigoens or not but Google has completely missed the point and employed the pigeons for ranking pages. This information about pigeons which Google ignore is a big threat to the core business as it depends on CPC / CPA. With the massive amount of data available to Facebook and the possible integrations with Bing, it will not take them much longer to come up with Adworks competition with better targeting. I will be surprised if Facebook doesn’t launch advertising outside of their massive platform via facebook share like widgets.

 

I am not impressed about Google+ expect for the added privacy features but that comes with a price called circles which are extremely difficult to create and manage even using a 22″ monitor. Also until today, I mainted different email ids and accounts for managing social networks and communication. Suddenly, combining social network to my primary e-mail id itself is the biggest privacy issuse as far as I am concerned.

But  Google will eventually learn about what we think about pigeons and over come the competition. Reason being the new doors which google has opened with the new product and upgrades to google adwords, Google apps etc. Its common knowledge that the social networks are the biggest consumers of content on portals like YouTube. With the recent announcement of DRM based YouTube rentals and availability of Social Gold reward coins I am assuming Google will be able to do some crazy things. Google Offers another very promising new launch which can be piggy backed on G+. A very cool idea Google Offers can try is ability to exchange Google’s virtual currency for offers via the new service. Game developers will be very soon opened doors to develop games on G+ & armed with the virtual currency Google has finally a weapon against Facebook’s plans to become world’s biggest bank. I am looking forward to see the roll out of location aware offers via G+ which can be further incentives by SocialGold and tieup all this with Google’s primary business of display advertisements. I am not forgetting about Google TV and building an extension to adwords which can show ads on TV and make people share it on G+ and much more. Groups is another massive amount of untapped data which G+ could make use of and from what I have read, the Google groups team is already sitting in the same building as that of the Emerald Sea building and cooking things. Looks like G+ is the technology giants attempt, a well rehearsed and executed attempt against threats from Bing+Yahoo & all knowing Facebook. Hope it works out and internet remains a free place rather than limited as pages within someone’s data center.

OpenID

Finally a decentralized way of signing into multiple services is here. Now with Yahoo and Google supporting OpenID looks like it is going to take over the world and replace other authentication systems. (good by .NET passport !)
I was using my OpenID URI http://freebird.myopenid.com for sometime now & wanted to use my domain as my OpenID.

how to use blog URL (blogspot) or a domain as OpenID URL

  1. Sign up for an OpenID with a ID provider. (http://myopenid.com is what I am using)
  2. open the index.html or template of your blog / website and add the following to head section.

<link rel=”openid.server” xhref=”http://freebird.in/wp-content/uploads/2008/02/server”  />

<link rel=”openid.delegate” xhref=”http://freebird.in/wp-content/uploads/2008/02/freebird.myopenid1.com”  />

The advantage of this method is that our OpenID will remain the same and we can change the ID provider.

In the case of a wordpress blog, we can edit the header.php file of the active template and add the above in the head section.

Enabling OpenID for wordpress blog

  1. get OpenID plugin for wordpress from here
  2. follow the instructions to activate the plugin
  3. if needed, edit the comments.php file of your template as mentioned the README file

Stealing Yahoo! login details

There is a URL http://www.geocities.com/hott_new_pics_for_you17/ floating around for some days which steals yahoo ids of users.

The login form in the page POSTS the form data to ” http://www2.fiberbit.net/form/mailto.cgi” This cgi script sends the Username and Password to the mail id : [email protected] in the following format.

____________________________

送信元ホスト名: MY_ISP / IP
送信元ブラウザ: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv: Gecko/20060728 Firefox/1.5.0.6
送信元URL :login = fgfg
passwd = gfgg
.save = Sign In
____________________________

The following mail id occurs when there is an error in form sumbission: [email protected]

pick of the day

Konfabulator Yahoo! widgets. Found this while searching for “flickr widget”. (hey thanks for mentioning about the widget pal)

Get it here: http://widgets.yahoo.com/

Desktop without the Widget (its a slow win box with 2.8GZ CPU and 1GB RAM. Sad. But Can’t help it)

voilà ! here is the desktop with the widgets, Two of them fetching photos from flickr, third one replacing the post-it notes.

The Bigger widget with tiled images is wickr Seems down but the page is there in Google cache