Wednesday, September 29, 2010

Yes, open source is a win-win

Open source is a win-win.

I am very pleased to let the MaraDNS community know that I have successfully gotten a job offer which I just accepted this week. This is a job offer which I would not have gotten if I had not kept my skills current while living in Mexico by continuing to develop MaraDNS, culminating with a three-year-long complete rewrite of the recursive code.

The timing for getting this offer is perfect. As I planned to do a year ago, now that MaraDNS 2.0 is out the door, I will put MaraDNS on the back burner. This will give me more time to be a productive worker at my new job and refine my programming knowledge to master new technologies, keeping my skills current in the 2010s.

I am going to continue to support MaraDNS, of course. I will fix important bugs brought up on the mailing list, and will naturally fix any critical security bug that may be found in MaraDNS. I will also answer emails on the mailing list when and if I have time to do so. But, I do not have any plans to add new features to MaraDNS at this time.

I would like to thank all MaraDNS users for all of your comments, feedback, bug reports, and feature requests. I have devoted a lot of the last decade to MaraDNS, and it has been an honor to freely give to the world something which I hope people find useful.

Open source development truly pays off. It’s a win for users, since they have good code which they can change any way they see fit; it’s a win for developers, because it allows them to showcase their work and demonstrate their programming discipline to prospective employers. I encourage people who are looking to improve their programming skills to become part of an open-source project. It can be frustrating at times; but it is also rewarding.

Tuesday, September 28, 2010

MaraDNS 2.0.01 released

I have released MaraDNS 2.0.01. This is simply MaraDNS 1.4 with the deprecated MaraDNS 1.x recursive DNS server removed, and with the documentation updated to reflect the fact that Deadwood 3 is now used for MaraDNS’ recursion. People, of course, are free to use MaraDNS 1.4 for as long as they want, but its recursive code is now deprecated and only supported for critical security updates.

I encourage users of MaraDNS to use Deadwood for all recursion with MaraDNS.

Sunday, September 26, 2010

Friday, September 24, 2010

Deadwood 3.0.01 released

I have released Deadwood 3.0.01 today. With this release made, no new features will be added to the Deadwood recursive resolver for the foreseeable future. However, bug fixes and other routine maintenance will still be applied to this software package.

Deadwood is now a stable recursive resolver suitable for embedded systems, desktop PCs, or on internet servers. Deadwood is fully 64-bit compatible, has optional support for IPv6, and does not require threads to resolve records.

Deadwood 3.0.01 will be the recursive resolver for MaraDNS 2.0, which I will release shortly, as well as being an optional recursive DNS server to use in MaraDNS 1.4.

It has been a pleasure developing Deadwood and MaraDNS for nearly a decade. I will continue to support MaraDNS and Deadwood but have no plans to add new features once MaraDNS 2.0 is released.

In addition, I no longer need to accept donations for MaraDNS. All Deadwood and MaraDNS support needs to be done via the MaraDNS mailing list.

The 3.0.01 release of Deadwood is available here:

Wednesday, September 22, 2010

New Deadwood snapshot: Parser bugs fixed

I have been uploading new Deadwood snapshots to fix some bugs in the dwood3rc parser; the parser now has the += (append) operator work correctly with dictionary variables, as well as having more stringent controls for dictionary indexes (dictionary variables must be initialized; a dictionary index can not be appended to without first being set; etc.) so that dwood3rc files must be working Python 2 scripts to correctly parse.

In addition, I have refined the documentation a little.

They can be looked at here:

Monday, September 20, 2010

New Deadwood snapshot: Two questions and answers added to FAQ

I have uploaded a Deadwood snapshot today that adds two questions and answers to the FAQ. One is about blacklisting domains with Deadwood, and the other is a brief note about how, as much as I would like Deadwood to have DNSSEC support, I am not in a place in my life where I have the time to implement DNSSEC “for fun and for free”.

The updated FAQ can be looked at here:

Thursday, September 16, 2010 is a scam

For people looking for work, beware of I got hit by their scam when I was looking for work last week. Based on the various reviews about this company out there about this company, I can safely say that they are a scam.

The scam works like this: has a number of job postings on their web site; once you sign up with them, they quickly send you an email that they have looked at your resume. They then tell you that your resume sucks, and that you should pay them $400 to improve your resume to have a chance in the job market.

They give this form letter to everyone who signs up for one of their postings on their site, giving out a letter like this: “Your resume does not pass the 30-second test, and the content is not up to the standards one would expect from a candidate like you. Countless studies have proven that resume quality is the key determinant as to whether a candidate is selected to be interviewed. Your resume needs a boost from a visual, content, and overall writing standpoint to engage the reader. It needs to make them want to learn more about you. I didn’t find it to be exciting, and it didn’t make me want to run to the phone to call you. In short, your resume is effectively sabotaging your job search.”

It does not matter how well-written one’s resume is. It does not matter if one has the best resume that was ever made. After has their spambot scan your resume, they give you the above scathing criticism. Indeed, according to one anonymous poster, resumes written by’s own resume writers get the above very negative review.

Googling the exact above phrase gives multiple results. Why am I not surprised?