Well, my recent posts here are basically about LDP, understandably it is the only topic in my work that keep me focused on that particular project. Since I saw the wiki would die I tried to do anything I could to save it. And there is some result, the spam abuse was decreased to “normal” and reasonable level. In order to achieve athat I put as much as possible efforts to keep the wiki’s content db up-to-date. But probably it is not the only thing necessary to protect the wiki. I consider a per-registration account, so any no-registered users will have disabled access to write and edit pages.

Since, this is a documentation project, I suppose the wiki to be a destination point for serious people but not spammers and hackers. And I hope a little registration should not annoy and disappoint possible contributor of documentation .

I am considering a weekly issues of community effort. Something like a place where all we could have a beer for a while and talk about the LDP. This is good idea, the only difficulty could be that we do not know each other. But that is the good part of hte internet, it connect people from many places around the world in a few seconds. A nice place for weekly meetings for beer and talking about the LDP could be the IRC. Good place where all we could talk and have a cold beer.

And, of course, LDP forums/blogs would be appropriate idea also for the people that are not available at the meeting’s time.

Basically I always thought that single and strong effort on converting from html to wiki is good enough to power-up the wiki. But the recent days gave me other thoughts. A closer look at tlkd.org show almost 500 howtos and that is too much for a single man with limited time and knowledge. And I would say that all the additional work is needed because of the errors made by the actual parser. So, no extra work is required by default but since the parser of html to wiki is not perfect one that extra work is needed for successful and appropriate conversion.

Actually I have 2 ideas for html to wiki convertion, the first one is the perl utility perl2wiki and second one is just Open Office. From what I have seen they both make mistakes and they both are not perfect so their produced files need extra work. Recently I got experience with one great thing in linux – flex and bison. They could possibly fix all that problems and probably produce effective working wiki text file ready for importation to the wiki.Delete post

LDP, the Linux documentation project is in strange position right now. I would clasify the it as dead or almost dead. Too bad for such a nice project but all things at this world have end some time. So, probably this is the time for LDP. But actually not this time.

Yeah the project is in trouble, because of the lack of contributors, authors, editors, lack of everything and everyone. I have no idea how this project manage to survive so  long until now. Probably there was a time when it was active but at the current moment the activity is equal to zero.

Most of the reasons responsible for the project failure are no coordination and weak organization. The whole model of work is weak, and not just that. There is huge number of documents and materials, howtos, tutorials and other stuff that just sit and take dust. That documentation is getting older and older even now at this second. That means at every spent moment that documentation becomes inappropriate for the current version of the software.  Basically I told my opinion about the work at LDP and now we try to build wiki that for the moment is continuously spammed by some very aggressive spammers.

How bad, the “spammer” today is not just a hobby but a job. In most cases even very well paid job. So, if we want to create that wiki a reliable protection is needed. So, I would clasify the following work needed to be done in LDP:

• Spam protection at Wiki
• Importing of all the documentation
• probably new web site, the one now is too old

At that point I would remark that already 5 or maybe 6 HOWTOs are imported to the wiki by me. So, every day I try to import 1 or 2 howtos and for the last 3 days I have imported 5,6 howtos. Not so great number but after 1,2 week there should be 40-50 howtos.

We all see what is happening in US and many countries around the world. Inflation, deflation, stagnation that are the normal processes now and even the US $ is falling down. But how this will hurt the open source companies? I think they all actually will benefit from this situation. While the closed source companies earn their money from selling of licenses the open source companies sell services and support. So, in that way the client is probably more willing to take a qualified support instead to buy a license.

But the real danger to all closed source companies is the economic crisis. Such a crisis reflects immediately to the client’s pocket and that reflects to the company. So, a company that make open source product has lower expenses because of the support and contributions of the community. For example if there are two operating systems on the market, a closed one and an open one. The close one has price of $200 and the open about $50 guess what will choose the customer.

So, in future the market will be more friendly and more friendly to the open source companies because of the cheaper products they could offer. But let’s take a look to the nowadays. Why GNU/Linux and friend are not big guns in the desktop market? Because of the advertisement. I have a look about some of the paid distributions and I think they all have lack of advertisement to the mass customer. They just do not advertise their products enough. That’s it. So, I would say “Keep working all you open source companies! You may have a difficult time now but the future is yours.”

And that is not all. There is some main target markets for the open source companies – support and services for a particular distribution, applications and selling gadgets. The “gadgets” is business is not so serious but it definitely earns some money. So, I will take a look about the support and services for a distribution and application. The user base is growing fast and if 1 year ago the gnu/linux users were 1000, so today they are 1500 (for example of course, we all know they are many thousands if not millions). That means the market for the above 2 areas is growing too. And if the other closed-source companies have difficult moment because of the crisis that is not so important for the open source companies. They have growing user base and that means growing market and more money for them. Lucky you, open source firms!

What a professional may look in one operating system?

Hmmm. It is not so easy to achieve this goals. Everybody is aware of the danger of the viruses and all the malicious software that wants to hurt your computer and everybody is aware how hard is to do it.

So, just think how many resources it costs to have an antivirus software with”some” effectiveness against the viruses.Think about the damage that a possible virus can do to the system as well as installed software and important documents. And what if you loose this documents? Not good, definitely.
But what if you can avoid all these unwanted troubles and danger? Did you think how you can save your time and money? And probably you can earn more money just because of improvement of the work.

ok. There is solution for all these points.

Many people have fear of GNU/Linux and everything that have relation with it. But now I will show some ways how one can improve the work and benefit in one shot. And all of this will be achieved via GNU/Linux. ok. How a distribution can improve anyone’s work? Very simple – via avoiding troubles. So, there is a distribution known as one of the easiest in the world. That is Freespire. Freespire is Ubuntu-based distribution made for the home user. Take a look about it. If this GNU/Linux can satisfy your needs then probably is more profitable to you to take the opportunity instead of working in the old way.

Freespire home page

Software repository

Community forums

At the first time when I started to use Freespire I was so surprised. Version 2.0.8 is so well-designed and everything worked. Ehh, alomst everything. With the using of the system I found out that my installation is progressively failing. I do not know why. But the effect and defect of this was only to the CNR. Every time I tried to install something I got the message “CNR received dpkg error”.

Then my next actions was to investigate the reason this error happen. I have used the tool “strace” and the following log files show all the actions of the CNR client.

strace stderr cnr output

strace stdout cnr output

After the first investigation for this error I found how it is happened. It seems the cause of this was the CNR client itself. When the client have failed with installation of particular application then next installations somehow reports error but in fact there is no error and the application is installed correctly. For me, that is obvious bug in the CNR client appears in some particular cases. My case was VirtualBox but probably it may happen with other applications too.

To fix it and make the client work properly I just removed the virtualbox package manually via apt-get. After that CNR client began to work as it has to.

I have pretty high interest to the all existing filesystems, especially the ones used in unix-like operating systems. So, recently my interested got higher and higher and now I have prepared a series of benchmarks to see the characteristics of some the todays popular filesystems like Ext2, Ext3, XFS, JFS, ReiserFS.

Ext2 filesystem log from tiotest benchmark

Ext3 filesystem log from tiotest benchmark

XFS filesystem log from tiotest benchmark

Reiserfs log from tiotest benchmark